University of Wollongong Research Online

University of Wollongong Thesis Collection University of Wollongong Thesis Collections

2002 Study of Clusterin : an extracellular Stephen Poon University of Wollongong

Recommended Citation Poon, Stephen, Study of Clusterin : an extracellular chaperone protein, Doctor of Philosophy thesis, Department of Biological Sciences, University of Wollongong, 2002. http://ro.uow.edu.au/theses/1056

Research Online is the open access institutional repository for the University of Wollongong. For further information contact the UOW Library: [email protected]

STUDY OF CLUSTERIN: AN EXTRACELLULAR CHAPERONE PROTEIN.

A thesis submitted in fulfilment of the requirements for the award of the degree of

Doctor of Philosophy

From

THE UNIVERSITY OF WOLLONGONG

By

Stephen Poon (B.Sc. Hons)

Supervisors - Mark R. Wilson* and John A. Carver^

Department of Biological Sciences t Department of Chemistry

2002 DECLARATION OF AUTHENTICITY

This thesis is submitted in accordance with the regulations of the University of Wollongong in fulfillment of the requirements for the degree of Doctor of Philosophy (PhD). It does not incorporate any material previously published or written by any person except where due reference is made in the text. The experimental work described in this thesis is original work and has not been previously submitted for a degree or diploma at any other university.

STEPHEN POON

11

\ ACKNOWLEDGEMENTS

I wish to acknowledge, with the utmost appreciation and thanks, the following who have shown great professionalism, patience and guidance over the course of my post-graduate degree: Associate Professor Mark R. Wilson, Department of Biological Sciences; and

Associate Professor John A. Carver, Department of Chemistry; both of University of

Wollongong, Wollongong, NSW, Australia. Without them, this thesis would not have been accomplished. I would also like to express my thanks and gratitude to Professor Martin

Tenniswood for putting up with me during an unforgettable term at the University of Notre

Dame, Indiana, USA.

I would also like to extend my thanks to my parents for their love and continual support throughout my degree. They encouraged me to aim high, to jump over every hurdle, break through all barriers, and to strive to be the best! Thanks for such invaluable advice.

A special thanks also to my mentor, Dr. David Humphreys (aka, "clusterin man"), whose guidance during my Honours and parts of Post-Graduate degrees have been momentous; and to Robyn A. Lindner for her forthcoming support. Finally, I would like to thank all my fellow students (past and present) for their friendship and understanding. In particularly, I would like to acknowledge Alison Smail, Rachel Jones, Christine Gillen, and Teresa

Treweek, for their jovial behaviour; without you guys, my time at Wollongong would have been very boring. I wish you all the best for the future and thanks for the memories. To

Elise Stewart and Justin Yulbury, keep pushing the frontiers of clusterin research forward.

I expect to see you publishing at least one paper a year!

111

^x Finally, a special mention of thanks to those with whom I have had the pleasure to be acquainted whilst at Notre Dame. To Kerry Gilmore, for letting me stay with you; to

Louise Flanagan, for being such a great friend, and to the following for accepting me into their 'Irish/Canadian/American community'; Lorna, Mark, Edmund, Gerry, Kathryn,

Kenneth, Soma, Sharon, Glendon, Judy, Sarra, Emma, Kathleen, and Uwe. Love you all!

STEPHEN POON

IV

\ ABSTRACT

Clusterin is a widely distributed and highly conserved secreted mammalian whose elevated expression has been detected in a number of states (e.g.

Scrapie, Alzheimer's and Creutzfeldt-Jakob ) that are associated with abnormally high levels of misfolded and/or precipitated . Clusterin, present at concentrations ranging from 35-105 ug/ml in human serum, is known to interact with a variety of proteins and in vitro. These numerous binding interactions have led to a number of biological functions being proposed for clusterin; these include roles in reproduction, transport, endocrine , , and complement regulation. In a recent study, it was reported that clusterin has molecular chaperone activity in vitro. Molecular chaperones, as defined by

John Ellis, are proteins that function in vivo to specifically interact with and stabilise unfolded or partially unfolded proteins, thereby preventing them from potentially aggregating and precipitating, e.g. during .

In this study, clusterin was shown to prevent the precipitation of heat-stressed ovotransferrin and y-, as well as DTT-reduced ovotransferrin and lysozyme.

Analysis by size exclusion chromatography of samples in which clusterin was co-incubated with ovotransferrin or lysozyme undergoing stress-induced denaturation revealed the presence of HMW species in the void volume that was eluted from the column. Subsequent analysis by SDS-PAGE of these HMW species, confirmed the presence of both clusterin and the stressed target protein. Results presented in this report also demonstrate that clusterin protects proteins in (i) diluted human serum from heat-induced precipitation and (ii) undiluted human serum from DTT-mediated precipitation. Other results presented in this thesis indicate that clusterin does not have the ability to hydrolyse ATP and hence, performs

V its chaperone function in an ATP-independent manner. In addition, clusterin was unable to independently facilitate the reactivation of heat-inactivated ADH and catalase after removal of stress. However, in the presence of a chaperone with refolding capability (i.e. Hsc70) and

ATP, clusterin-stabilised ADH and catalase were partially refolded. Taken together, these results raise the possibility that clusterin may inhibit precipitation of human serum proteins in vivo and create a reservoir of inactive protein structures from which folding-competent proteins can be subsequently reactivated by other refolding chaperones.

Clusterin was shown to inhibit the slow precipitation of y-crystallin and lysozyme, but was unable to prevent these same target proteins from rapid precipitation. Real-time H

NMR spectroscopic analysis of the interaction between clusterin and a-lactalbumin reveal that clusterin did not alter the rate of a-lactalbumin reduction but did stabilise the less ordered intermediately folded form of the protein. These results suggest that (i) kinetic factors are important in the chaperone action of clusterin and (ii) clusterin binds specifically to slowly aggregating proteins on the irreversible off-folding pathway.

This thesis also presents results to show that elevated temperature (up to 50 °C) does not induce significant changes in the oligomerisation state of clusterin nor does it substantially alter the ability of clusterin to interact with heat-stressed or chemically reduced target proteins. In contrast, incubation at mildly acidic conditions resulted in the dissociation of clusterin oligomers, which led to an increased exposure of hydrophobic regions on clusterin to solution and a concomitant enhancement of its chaperone action. A phenomenon known as acidosis occurs at sites of tissue damage or where the local pH can drop below 6. Acidosis has been reported to occur at sites of inflammation as well as in many of the diseases to which clusterin has been associated. Taken together, these results

vi suggest that under these conditions, the dissociation and enhanced chaperone actions of clusterin could help to inhibit the aggregation and deposition of inflammatory and/or toxic insoluble protein deposits which would otherwise exacerbate pathology.

At present, the structural regions responsible for the ligand binding and chaperone action of clusterin have not been identified. Sequence analysis of clusterin has revealed several regions that could be functionally important, including three regions of amphipathic a-helices and two coiled-coil domains. In addition, studies have shown that, despite having variable truncations at the C-terminus of the a-chain and the N-terminus of the (3-chain, clusterin expressed by yeast Pichia pastoris has similar chaperone activity to human clusterin in vitro, indicating that the sites responsible for the chaperone action of clusterin are more likely to be located more towards the N-terminal region of the a-chain and the C- terminal region of the [3-chain. In order to identify the functional sites of clusterin and to test the above statements, five proline-substitution and five truncation clusterin mutants, as well as wild-type clusterin, were developed. These mutants, including wild type clusterin, were expressed in transiently transfected Spodoptera frugiperda (Sf9) insect cells. Sf9 cells were chosen as the host for the expression of wild type and mutant clusterin due to their reported ability to express high levels of recombinant proteins and perform post-translational modifications in a manner that is similar to mammalian cells. However, Western blot analysis showed that the expressed proteins were not expressed at high levels and had molecular masses that were approximately 15 kDa smaller than their expected sizes. Since the Sf9 transfectants were not cloned, the Sf9 cultures may have contained a large number of non-transformants/non-secretors that may have outnumbered secreting transformants and hence, explain why the yield was so low. The size irregularity can be

vu explained by incomplete of the expressed proteins or by truncations caused by protease actions. Whatever the case may be, these results clearly indicate that Sf9 insect cells are unsuitable for producing properly processed wild-type or mutant clusterin.

Therefore, to produce properly processed wild-type and mutated clusterin for the study of clusterin's structure-function relationship, a different expression system will be required.

Vlll TABLE OF CONTENTS

TITLE PAGE

DECLARATION

ACKNOWLEDGEMENTS in

ABSTRACT

TABLE OF CONTENTS IX

ABBREVIATIONS XV

LIST OF TABLES XV111

LIST OF FIGURES XIX

1. INTRODUCTION 1

1.1 General Introduction 2

1.2 Proteins 2

1.2.1. Protein Structure and Function 2

1.2.2. Protein Folding 7

1.2.3. Protein Unfolding/Misfolding and its Consequences 10

1.2.4. Molecular Chaperones 13

1.3 Clusterin 15

1.3.1. Clusterin Structure and Biosynthesis 16

1.3.2. Clusterin Distribution and Binding Interactions 22

1.3.3. Proposed Biological Functions of Clusterin 24

1.3.3.1 Clusterin and Reproduction 25

1.3.3.2. Involvement of Clusterin in Lipid Transport 26

1.3.3.3. Clusterin and Apoptosis 27

1.3.3.4. Clusterin in Complement Regulation 31

IX 1.3.3.5. Clusterin as a Molecular Chaperone

1.4 Thesis Objective 35

GENERAL MATERIALS AND METHODS 36

2.1 General buffers and solutions 37

2.2 SDS-Polyacrylamide Gel Electrophoresis (SDS-PAGE) 37

2.2.1 SDS-PAGE Buffers 37

2.2.2 SDS-PAGE Methods 38

2.2.2.1 Loading Protein Samples and Performing

Electrophoresis 38

2.2.1.2 Staining and Destaining 38

2.2.1.3 Gel Drying and Storage 38

2.3 Agarose Gel Electrophoresis 39

2.3.1 Buffers for Native and DNA Agarose Gels 39

2.3.2 Agarose Gel Electrophoresis Gel Set-up 40

2.3.3 Agarose Gel Electrophoresis Methods 40

2.3.3.1 Casting and Running Agarose Gels 40

2.3.3.2 Visualisation of DNA or Protein Bands 40

2.4 Western (Protein) Transfer 41

2.5 Immuno-Labelling of Membrane-Bound Proteins 41

2.6 Enhanced ChemiLuminescence Detection (ECL) 44

2.7 Immuno - 'Dot Blots' 44

2.8 BCA Micro-Protein Assay 45

2.8.1 Methods 45

2.9 Immunoaffinity Purification of Clusterin 46 2.9.1 Preparation of Human Serum 46

2.9.2 G7-Sepharose Purification of Clusterin 46

2.9.3 Protein G Purification of Clusterin 47

3. CLUSTERIN HAS CHAPERONE ACTIVITY SIMILAR TO THE SMALL HEAT-SHOCK PROTEINS 51

3.1. Introduction 52

3.2. Methods 58

3.2.1. Inhibition of Protein Precipitation 58

3.2.2. ELISA 59

3.2.3. Sephacryl S300 Size Exclusion Chromatography 60

3.2.4. SDS-PAGE Analysis of Clusterin-Substrate Complexes 61

3.2.5. Assays for Heat-Induced Loss of Enzyme Activity 61

3.2.6. ATPase Assays 62

3.2.7. Thermal Denaturation and Renaturation Experiments 63

3.3. Results 65

3.3.1. Clusterin Maintains the Solubility of Stressed

Ovotransferrin, Lysozyme and y-Crystallin 65

3.3.2. Clusterin Protects Serum Proteins

from Stress-Induced Precipitation 67

3.3.3. Clusterin Binds Preferentially to

Stressed Forms of Ovotransferrin and Lysozyme 69

3.3.4. Clusterin Interacts With Stressed Ovotransferrin, Lysozyme,

and a-Lactalbumin to Form HMW Complexes 71

3.3.5. ATP Does Not Affect the Ability of Clusterin

XI to Suppress Stress-Induced Protein Aggregation 72

3.3.6. Clusterin Does Not Protect Enzymes

from Stress-Induced Loss of Function 76

3.3.7. Clusterin Has no Detectable ATPase Activity 77

3.3.8 Clusterin is Unable to Independently Promote Reactivation

of its Substrates but it can Stabilise Proteins in a

State Competent for Refolding by Hsc70 78

3.4. Discussion 81

4. CLUSTERIN IS A pH-DEPENDENT CHAPERONE WHICH SPECIFICALLY INTERACTS WITH DISORDERED MOLTEN GLOBULE STATES OF PROTEINS 88

4.1. Introduction 89

4.2. Methods 94

4.2.1 Size-Exclusion Chromatography 94

4.2.2 ELISA 94

4.2.3 Protein Precipitation Assays 95

4.2.4 NMR Spectroscopic Analysis of Interactions

Between Clusterin and Reduced a-Lactalbumin 98

4.3. Results 99

4.3.1 Temperature Has Negligible Effect

on the Oligomerisation State of Clusterin 99

4.3.2 Elevated Temperature Slightly Enhances

Substrate Binding by Clusterin 100

4.3.3 Chaperone Action of Clusterin is only Slightly

Xll Enhanced at Elevated Temperature 100

4.3.4 Low pH Enhances Binding of Clusterin to Stressed Proteins 103

4.3.5 Low pH Enhances Clusterin-Mediated

Inhibition of Protein Precipitation 105

4.3.6 Clusterin Does Not Inhibit the Rapid Precipitation

of Lysozyme or y-Crystallin 108

4.3.7 Real-Time ]H NMR Analysis of the Interaction

Between Clusterin and Reduced a-Lactalbumin 110

4.4 Discussion 114

5. EXPRESSION OF RECOMBINANT CLUSTERIN TO INVESTIGATE THE STRUCTURE-FUNCTION RELATIONSHIP OF WILD-TYPE HUMAN CLUSTERIN 120

5.1 Introduction 121

5.2 Methods 128

5.2.1 Determining the Sites for Proline

Substitution or Truncation in Human Clusterin 128

5.2.2 Quickchange™ Site-Directed Mutagenesis (Overview) 129

5.2.3 Transforming into XL 1 -Blue Supercompetent Cells 134

5.2.4 Gateway™ Cloning Technology (Overview) 134

5.2.4.1 Generation of attB-Flanked PCR Products 135

5.2.4.2 Creating Entry Clones Via the BP Reaction 140

5.2.4.3 Creating Expression Clones Via the LR Reaction 142

5.2.5 Transfection of Sf9 Insect Cells 143

5.2.6 Wizard® Plus Miniprep DNA Purification System 144

Xlll 5.3. Results 145

5.3.1. Chosen Sites for Proline-Substitution and Truncation in Human Clusterin 145

5.3.2 Site-Directed Mutagenesis 148

5.3.3 PCR Amplification of Wild-type and Mutant

Clusterin and Creation of Entry Clones 148

5.3.4 Creation and Analysis of Expression Clones 150

5.3.5 Expression of Wild-type and Mutant Clusterin in Sf9 Cells 151

5.4. Discussion 155

GENERAL CONCLUSIONS 160

REFERENCES 170

APPENDIX 194 ABBREVIATIONS

3D Three Dimension a-Crys Alpha-Crystallin a-Lac Alpha-Lactalbumin

A210, A28o or A350 Absorbance reading at 210, 280 or 350 nm Ab Antibody ADH Alcohol Dehydrogenase AK Adenylate Kinase AMP Adenosine Mono-Phosphate ANS 8-Anilino-l-Naphtalene Sulfonate Apo A-I A-I ATP Adenosine Tri-Phosphate

Az Azide (N3) BSA Bovine Serum Albumin Cat Catalase CD Circular Dichroism CDS Clusterin-Depleted Serum CJD Creutzfeldt-Jakob Disease CL Clusterin cm Centimeter dH20 Distilled Water DMEM Dulbecco's Modification of Eagle's Medium DNA Deoxyribonucleic Acid DTT Dithiothreitol E. coli Escherichia coli EDTA Ethylenediamine Tetraacetic Acid ELISA Enzyme-Linked Immunosorbent Assay EtBr Ethidium Bromide FCS Foetal Calf Serum

XV FITC Fluorescein Isothiocyanate

FL Fluorescence y-Crys Gamma-Crystallin

GST Glutathione-S-Transferase GuHCl Guanidine Hydrochloride

HDC Heat Denatured Casein

HDL High Density Lipoprotein

HMW High Molecular Weight HRP Horseradish Peroxidase

Hsp

Hsc Constitutively expressed heat-shock protein

IAA 5 mM R(+)-[6,7-dichloro-2-cyclopentyl-2,3-dihydro-2-

methyl-l-oxo-lH-inden-5-yl)oxy] acetic acid Ig Immunoglobulin kDa KiloDaltons LDH Lactate Dehydrogenase LRP-2 Lipoprotein Receptor-related Protein-2

Lys Lysozyme MAb Monoclonal Antibody

MDH Malate Dehydrogenase mg milligram (1 x 103 g)

MG Molten Globule

(J.g microgram (1 x 10"6 g) mL millilitre (1 x 10"3 L)

(iL microlitre(l x 10"6L) nm Nanometer

NMR Nuclear Magnetic Resonance

NS Normal Human Serum

Ovo Ovotransferrin

PCD

PI Propidium Iodide

XVI PBS Phosphate Buffered Saline PVDF Polyvinylidene Difluoride rpm Revolutions per minute RO Reversed Osmosis RTF Ram rete Testis Fluid SDS-PAGE Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis

Sf9 Spodoptera frugiperda IPLB-Sf9 sHsps small Heat Shock Proteins SMR Subunit Molar Ratio TEMED N,N,N' ,N' -Tetramethyl-Ethylenediamine TGFp Transforming Growth Factor P TNFa Tumour Necrosis factor a V Volts v/v Volume/Volume w/v Weight/Volume WT Wild type

XVll LIST OF TABLES

Table 1.1 Tabulated summary of the names, origins and the year of discovery of common clusterin homologues. 17

Table 2.1 Compositions of the buffers and stains used during SDS-PAGE. 37

Table 2.2 Compositions of the various reagents used during native agarose gel electrophoresis. 39

Table 2.3 Compositions of the various reagents used during DNA agarose gel electrophoresis. 39

Table 2.4 Compositions of BCA reagents used for the determination of protein concentrations. 45

Table 2.5 A complete list of chemicals and reagents (including their formulae and the name of their supplier) used throughout this thesis project. 49

Table 3.4.1 Data for the various protein substrates for which clusterin has been shown to inhibit stress-induced precipitation. 84

Table 5.1.1 Characteristics of the various clusterin mutants developed in this study. 127

Table 5.2.1 List and amounts of various components required for Quickchange™ mutagenesis of the clusterin in the plasmid, pTZ-HT7. 133

Table 5.2.2 Cycling parameters for the PCR-based incorporation of proline residues into clusterin gene via Quickchange™ mutagenesis. 133

Table 5.2.3 Cycling parameters for PCR amplification of the clusterin gene flanked by attB sites. 140

Table 5.2.4 List and amounts of various components required for creating entry clones via the BP recombination reaction. 14)

Table 5.2.5 Components and their amounts required for creating expression clones for wild-type and mutant clusterin. 142

XVlll LIST OF FIGURES

Figure 1.1 Structural movements as a result of substrate binding to adenylate kinase 3 (AK3).

Figure 1.2 Schematic representation of the various protein-folding pathways.

Figure 1.3 A pictorial representation of the extent of molecular crowding by macromolecules in a bacterial cell. 11

Figure 1.4 Images of amyloid fibrils. 13

Figure 1.5 Microscopy images showing the "clustering" effects of clusterin on TM-4 cells. 16

Figure 1.6 Schematic representation of the structure of clusterin. 20

Figure 1.7 Comparison of the primary sequence of clusterin homologues from human, rat, canine, bovine, quail, and the partial sequence of hamster clusterin. 21

Figure 1.8 Schematic outline of the comparative morphological differences that occur in necrotic and apoptotic cells. 29

Figure 2.1 A Schematic representation of the 'mini gel holder cassette' for the Bio-Rad western transfer unit. 42

Figure 2. IB A schematic representation of the general setup for the Bio-Rad western transfer unit for transferring proteins from SDS-PAGE gels onto a membrane. 42

Figure 2.2 Protein (Western) transfer by capillary action. 43

XIX Figure 3.3.1 Precipitation of ovotransferrin, lysozyme, alcohol dehydrogenase (ADH) and y-Crystallin as a function of time during stress. 66

Figure 3.3.2 Effects of clusterin on heat-induced precipitation of Human serum proteins. 68

Figure 3.3.3 Results of ELISA measuring the binding of purified clusterin or clusterin in crude human serum to native or stressed adsorbed proteins. 70

Figure 3.3.4 Sephacryl S-300 size exclusion chromatographic detection of HMW complexes between clusterin and stressed ovotransferrin or lysozyme. 73

Figure 3.3.5 SDS-PAGE gel and immunoblots showing the presence of clusterin and stressed proteins in the HMW complexes obtained in the exclusion peaks from size exclusion chromatographic analyses. 74

Figure 3.3.6 Effects of ATP on the ability of clusterin to inhibit stressed-induced protein precipitation. 75

Figure 3.3.7 Inability of clusterin to prevent heat-induced loss of enzyme activity. 76

Figure 3.3.8 Spectrophotometric time course measurements showing that clusterin lacks measurable ATPase activity. 77

Figure 3.3.9 Post-stress refolding of clusterin-stabilised ADH and catalase by Hsc70. 79

Figure 4.1.1 A schematic representation of the protein folding/unfolding pathways. 91

Figure 4.3.1 Effects of temperature on the oligomerisation state of clusterin. 99

Figure 4.3.2 Results of ELISA measuring the temperature-dependent binding of clusterin to stressed proteins. 101

Figure 4.3.3 Effects of elevated temperature on clusterin's chaperone activity. 102

XX Figure 4.3.4 Effects of mildly acidic pH on the binding of clusterin to heat-stressed ovotransferrin, glutathione-S-transferase, and reduced ovotransferrin, a-lactalbumin, and bovine serum albumin (BSA). 104

Figure 4.3.5 Effects of mildly acidic pH on the binding of clusterin (in unfractionated human serum) to heat-stressed GST at pH 6.0, pH 6.5, or pH 7.5. 105

Figure 4.3.6 Effects of mildly acidic pH on clusterin's chaperone action. 106

Figure 4.3.7 Plot showing the effects of low pH on the chaperone activity of a-crystallin. 107

Figure 4.3.8 Effect of clusterin on the slow and rapid precipitation of y-crystallin and lysozyme. 109

Figure 4.3.9 Real-time 1H NMR spectra and kinetics of a-lactalbumin reduction in the presence or absence of clusterin. 111

Figure 5.1.1 Schematic representation showing the mutation sites in the clusterin sequence. 126

Figure 5.2.1 Overview of the QuickchangeTM site-directed mutagenesis method. 130

Figure 5.2.2 Map of various plasmid vectors that were used to create the GatewayTM expression clones. 132

Figure 5.2.3 General overview of the GatewayTM cloning system. 136

Figure 5.2.4 A schematic representation showing the versatility of the GatewayTM cloning technology. 137

Figure 5.2.5 Diagram depicting the development of aflB-flanked PCR products via two-stage PCR amplification using two sets of overlapping Gateway -specific oligonucleotide primers 138

XXI Figure 5.3.1 Image of an ethidium bromide stained 1% agarose gel electrophoretic analysis of the attB-PCR products. 149

Figure 5.3.2 Image of an ethidium bromide stained 1% agarose gel electrophoretic analysis of the pENTR (entry) clones. 150

Figure 5.3.3 Image of an ethidium bromide stained 1 % agarose gel electrophoretic analysis of the various expression clones. 151

Figure 5.3.4 Immuno-dot blot analysis of tissue culture supernatants obtained from transfected Sf9 cells two weeks after transfection. 152

Figure 5.3.5 Western blot of a 10% SDS-PAGE gel of purified human serum and Sf9-derived recombinant clusterin 153

Figure 6.1.1 A schematic representation showing the proposed mechanism of chaperone action of clusterin during protein unfolding. 164

XXll Chapter 1

GENERAL INTRODUCTION

The Role of Clusterin in Protein Unfolding and Refolding 1.1 GENERAL INTRODUCTION

In 1997, studies conducted by Humphreys et al. (the results were published in

1999) revealed that a secreted mammalian protein called clusterin exhibited chaperone- like activities in vitro similar to those of the small heat shock proteins (sHSPs)

(Humphreys et al, 1999). For the first time, an extracellular mammalian chaperone protein had been described. As a consequence of this remarkable discovery, current studies reported herein have been aimed at determining the extent to which clusterin is involved in protein unfolding and, especially, refolding. This chapter focuses on introducing the relationship between protein structure, function and folding, and the importance of chaperone proteins to these processes. In addition, the general properties of clusterin and the relationship of its molecular chaperone activity to in vivo protein unfolding and refolding will be discussed in detail.

1.2 PROTEINS

1.2.1 Protein Structure and Function

The secret to the versatility of proteins lies in their complex structure; that is, the way in which they are ultimately folded. Central to the study of protein structure and function is a dogma, which states that, "3D structure is taken to be an obligatory prerequisite for protein function" (Dunker et al, 2001). This protein structure-function relationship can be represented by the following as:

Amino Acid Sequence • 3-Dimensional Structure • Function

What this implies is that the shape and function of all proteins is inextricably linked in such a way that, in order to be biologically active, proteins must adopt a folded three-

2 dimensional structure that is specific for their intended function. Hence, a protein is rendered functionally useless if it is not folded correctly.

For nearly a century, scientists have sought to discover the underlying mechanisms by which proteins are transformed into their functionally active conformation in vivo once they are synthesised. Their success has been noted by a number of milestones; reviewed in (Dunker et al, 2001) and summarised as follow. In

1929, Hsien Wu postulated that native protein structures are formed by repeated patterns of folding into a complex three-dimensional shape. In 1931, Anson and Mirsky demonstrated that haemoglobin folding is reversible and that the renatured form exhibited a native-like oxygen binding and tryptic digestion pattern. In the 1950s, studies by

Eisenberg and Schwert demonstrated that thermodynamic factors control the processes of denaturation and renaturation, and that these processes resulted in large changes to the conformation of the proteins involved. Also in the 1950s, Linus Pauling discovered two simple but regular arrangements of amino acids (the a-helix and (3-sheet) that were present in almost every protein (Pauling and Corey, 1951). In 1961, Christian Anfinsen and his colleagues (Anfinsen and Haber, 1961) showed that denatured bovine pancreatic ribonuclease (RNase), with reduced disulfide bonds and disrupted tertiary structures, can spontaneously but slowly revert back to its biologically active structure in vitro. They had demonstrated for the first time that "no special genetic information, beyond that contained in the amino acid sequence, is required for the proper folding of a molecule and for the formation of 'correct' disulfide bonds" (Goldberger et al, 1963; Ruddon and

Bedows, 1997). These and many other scientists have contributed greatly to the current understanding of protein structure and folding. Their findings have been summarised as follows.

3 Briefly, protein structure can be broken down into four distinct levels, called primary, secondary, tertiary and quaternary structures; listed here in the hierarchal order of increasing structural complexity (van Holde, 1990). Ultimately, as deduced from

Anfinsen's classical RNase experiment (see above), the three-dimensional shape (tertiary or quaternary structure) of all proteins is determined by the primary structure; that is, the sequence of amino acids that make up a particular protein (Anfinsen and Haber, 1961).

All proteins, from bacteria to humans, are constructed from the same repertoire of 20 basic amino acids, each of which has its own unique chemical 'personality' as a direct result of variations in the molecular composition of their 'side chain' (reviewed in

Branden and Tooze, 1999). Some amino acids favour aqueous environments and are more likely to be located on the surface of a protein where they can interact with the surrounding water molecules, whereas those that prefer hydrophobic environments tend to be located deep within the protein. Some have slightly positive or negative charges and, therefore, have a greater propensity to repel groups of similar charge. Moreover, some have a rigid side chain and hence lack flexibility and free rotation about the peptide bonds, whilst others are bulky due to having long or large side chains. This spectrum of behaviour, in turn, affects how all amino acids are spatially oriented with respect to each other within a mature protein. More importantly, it dictates whether a segment of polypeptide becomes part of an a-helix, P-sheet or P-turn (secondary structural elements), as well as dictating the final orientations or topology of these domains in the three-dimensional (tertiary and quaternary) structure of a folded protein (Branden and

Tooze, 1999).

As a general rule, proteins can exist in one of three major thermodynamic states at any given moment; these include the native, intermediary "molten globule" (MG) and the unfolded states (Ohgushi and Wada, 1983). The basis for the classification of a particular

4 protein into one of these states lies with the amount of structure that it possesses, as outlined in Table 1.1. While it would be easier to describe proteins as discrete structures that exist as homogeneous entities in a given state, the fact is that each state is comprised of an ensemble of different protein conformations with similar properties (Pain, 2000).

Ultimately, all proteins try to fold into the native state - the conformation in which they are thermodynamically most stable and of the lowest energy. The driving force for this stability lies with favourable negative enthalpy arising from harmonious intramolecular side chain interactions (through electrostatic interactions, internal hydrogen bonding, or van der Waals interactions), and a decrease in structural randomness, resulting in favourable negative entropy (Branden and Tooze, 1999; Dill,

1990; Shirley, 1992). For globular proteins, favourable entropy also arises from the burial of hydrophobic groups within the protein, which effectively causes the cessation of the hydrophobic effect (Branden and Tooze, 1999; Herzfeld, 1991). Finally, for a number of proteins, structural stability may also arise from the formation of disulfide bonds between cysteine residues located either on the same polypeptide chain or on adjacent chains (Creighton, 1997).

It is, however, erroneous to assume that a fully folded protein in its native state and stabilised by the above-mentioned forces and bonds is static. Random secondary structural elements as well as whole domains continually undergo movements in space.

Such fluctuations can be caused by individual atoms rotating around their bond axis or by the collective movement of groups of atoms (Kim and Woodward, 1993). Usually, these fluctuations are small, a few tenths of an Angstrom, and occur on a picosecond timescale

(Pain, 2000). Occasionally, larger fluctuations can also occur and the regions that undergo such fluctuations are often referred to by crystallographers as "flexible regions".

Structural movements are very important for the function of many proteins, being

5 involved in the processes of enzyme catalysis, receptor-ligand interactions, binding of antigens to antibodies, energy transductions, and so on. To illustrate this point, consider the enzyme, AK3, a member of the adenylate kinase family. Figure 1.1 depicts the structures of substrate-free, inactive and activated AK3, with both substrates (AMP and

Ap5P) bound (Schulz, 1991).

A B

Figure 1.1: Structural movements as a result of substrate binding to adenylate kinase 3 (AK3). Only non- hydrogen atoms are shown. (A) AK3 with no ligand. The 30-residue segment forming the AMP-binding site is indicated within the red circle. Structural movement of this segment upon AMP binding is shown by the unfilled arrow. (B) AK3 with bound substrate AMP. (C) AMP-bound AK3 rotated by 90 ° around the vertical axis. The flexible 38-residue domain is indicated within the blue circle. The direction of its movement upon Ap5A binding is indicated by the unfilled arrow. (D) AK3 with bound substrates AMP and Ap5A. These images were taken from Shulz, 1991.

6 Substrate-free AK3 is characterised by the presence of a large cleft which forms the active site, a main body consisting of a parallel P-sheet structure surrounded by a number of a-helices, a 30-residue segment forming the AMP-binding site, and a highly flexible 38-residue domain situated at the mouth of the cleft (Dreusicke and Schulz,

1988). As the first substrate (AMP) binds, the main body remains intact while the 30- residue segment undergoes a movement of 8 A followed by another 8 A on binding of

ApsP. At this moment, the 38-residue domain at the mouth of the cleft rotates by as much as 90 ° to form a solid lid, thus shielding the enclosed active site and rendering the enzyme fully active to carry out its function. This is, of course, a reversible process, but nevertheless it does illustrate the significant structural movements required for function in certain proteins, as well as the importance of having correctly folded motifs to permit proteins to function properly.

1.2.2 Protein Folding

Much of the initial understanding of how proteins fold into the native state has been gained through structural characterisation of the various protein conformations that feature in the protein-folding process. Experimental approaches to these studies included the use of circular dichroism spectroscopy, X-ray crystallography and protein fluorescence. The structural information gained from these techniques have often been used, like pieces from a jigsaw puzzle, to piece together folding pathways. The advent of

Nuclear Magnetic Resonance (NMR) spectroscopy and recombinant DNA technology has allowed the protein-folding process to be further studied at the molecular level. For instance, the ability to measure thermodynamic changes of individual or groups of atoms over time using NMR spectroscopy has meant that it is now possible to determine the kinetics of protein folding (Balbach et al, 1996; Creighton, 1997). In addition,

7 recombinant DNA technology (see chapter 5) has enabled production of mutants to probe the folding process and to ascertain the level of involvement of individual amino acids.

Therefore, besides their ability to characterise various protein transition states, these powerful, complementary techniques have facilitated substantial elucidation of the kinetics and thermodynamics that govern protein folding.

As a result of these developments, a number of models have been proposed to describe the mechanisms and pathways by which unfolded polypeptide chains are folded into the stable native conformation. In the simplest model, which usually applies to small, single domain proteins (typically with less than 100 residues), the process of protein folding is best described as being "two-state"; involving a direct transition from the high energy unfolded form to a low energy native state (Figure 1.2 A) (Jackson and

Fersht, 1991). In contrast, folding by larger, more complex proteins occurs via a general pathway involving an initial collapse of the unfolded state into a partly structured, intermediate "molten globule" state, followed by further structural remodelling and packing into the final native form (Sali, 1994). For some proteins, such as chymotrypsin inhibitor 2 (CI2) and barnase, folding may be achieved through the course of a single major pathway (Figure 1.2 B) (Jackson and Fersht 1991; Fersht, 1993), whereas the folding of other proteins, such as lysozyme, may involve numerous but parallel folding pathways (Figure 1.2 C) (Fersht et al, 1994; Matthews, 1993).

During protein folding, the transition from the unfolded to the molten globule state is characterised by (i) a substantial development of secondary structural elements such as a- helices and P-sheets, (ii) an absence of much of the tertiary structures associated with the native conformation, and (iii) the formation of a loosely packed hydrophobic core that results in the exposure of hydrophobic surfaces to solvent. This transition is fast, usually

8 occurring in a few milliseconds. As proteins evolve from the molten globule to the native state, a gradual assembly of persistent native-like tertiary structure (in the form of subdomains) is observed. This process, mainly driven to completion by the same forces that govern native protein stability, occurs much more slowly and can last up to 1 second

(Anfinsen, 1972; Branden and Tooze, 1999; Dill, 1990; Sali, 1994).

Unfolded native (A) protein protein

Unfolded native •• intermediate (B) protein protein

intermediate,

Unfolded native -• intermediate, (C) protein protein

intermediate.

Figure 1.2: Schematic representation of the various protein-folding pathways. (A) Two-state model; small, single domain proteins (typically with less than 100 residues) fold directly from the unfolded state into the native state via this model. (B) and (C) Folding pathways involving protein intermediates; folding of some proteins (e.g. barnase) are achieved through the course of a single major pathway (B), whilst others (e.g. lysozyme) may fold via numerous but parallel pathways (C). The protein structures were drawn to indicate different conformational states; they do not represent any known protein structures. The number of possible folding pathways as shown in (C) may vary between proteins.

9 1.2.3 Protein Unfolding/Misfolding and its Consequences

Protein unfolding is essentially the opposite process to protein folding. It may involve a partial or full unravelling of the native protein structure to form the molten globule or the fully unfolded polypeptide, respectively. Triggers for protein unfolding vary significantly (see below) but the outcome is always the same - they invariably lead to the weakening of thermodynamic forces that normally stabilise the protein conformer.

The molten globule structure has been a focus of research since it is in this state that protein misfolding and aggregation are most prevalent. It is now widely believed that the highly exposed hydrophobic domains of molten globules are partly responsible for promoting inappropriate protein-protein interactions.

Proteins generally fold quickly, thus reducing the amount of time that is available for protein intermediates to interact with each other to form large, useless protein aggregates. Although all proteins are able to fold quickly, Anfinsen demonstrated that more complex, multidomain, or oligomeric proteins are unable to fold as efficiently as small, single domain proteins unless they are folded in the presence of additional proteins, which Laskey termed "molecular chaperones" (Laskey et al, 1978). John Ellis

(Ellis, 1987) later applied this term to include all bacterial and eukaryotic proteins that function to prevent illegitimate interactions between different polypeptide chains during normal protein folding or upon unfolding under altered physiological conditions or stress.

The term stress is often used to describe any conditions that may influence or lead to protein unfolding and misfolding. Stress can be exerted by a number of factors such as temperature, pH, oxidation or reduction, and molecular crowding. In addition to these stresses, protein misfolding can also result from defective genes that encode for the production of mutated proteins with altered structural characteristics and hence, defective folding abilities. Molecular crowding is a stress factor of particular importance because it

10 affects all proteins, from newly synthesised polypeptides to mature proteins. Molecular crowding occurs due to the high concentration of transiently hydrophobic and charged proteins and other molecular species within the cell. For instance, the total macromolecular concentration in the of a typical bacterium such as E.coli is approximately 340 mg/ml (Ellis and Haiti, 1996); this overcrowding is graphically depicted in Figure 1.3. Overcrowding of molecules promotes "improper interactions" between nascent proteins, which, like that caused by other stress conditions, often leads to irreversible formation of non-functional misfolded protein structures or aggregates

(Van den Berg et al, 1999). The breakdown of controlled or regulated protein folding often leads to cellular malfunctioning and disease (Thomas et al, 1995; Dobson, 1999).

\Sf^' T t m^L^L^^\^ir* 1 0**9 ) r1' o GroEL H*' Ribosome

WfrtX-S/ftm \ m Protein '-* \j4 L^u DNA Si RNA ^r^g^i^jj { rtii%"S

Figure 1.3: A pictorial representation of the extent of molecular crowding by macromolecules in a bacterial cell. The identity of these macromolecules are indicated to the right of the picture. This image was taken from Pain, 2000.

Many diseases have been linked to the failure of proteins to fold properly or, if folded, to remain in the native conformation. In most cases, the aggregation of proteins

11 into insoluble complexes has been a key characteristic feature of these diseases (reviewed in Kapito and Ron, 2000; Thomasson, 2001). Some of these diseases include cystic fibrosis, Alzheimer's disease, Parkinson's disease, cataract, and prion diseases such as scrapie, bovine spongiform encephalopathy (BSE) and Creutzfeldt-Jakob disease (CJD).

In the case of cystic fibrosis, a single deletion-mutation of a phenylalanine at position 508 in the cystic fibrosis transmembrane conductance regulator (CFTR) causes the protein to misfold, rendering it unable to regulate the transport of CI" ions in epithelial cells

(Ackerman and Clapham, 1997; Kleizen et al, 2000). Cataract, which accounts for 42% of blindness world-wide, occurs when the aggregation of crystallin proteins, the major component of the eye, becomes so severe that the lens goes opaque, causing cataract sufferers to lose their sight. The other diseases listed above all feature extensive accumulation of amyloid fibrils. Amyloid fibrils are formed as a result of inappropriate interactions between P-sheets of partially unfolded proteins that polymerise to form long, unbranched fibrous structures of indefinite lengths (Figure 1.4). In the case of

Alzheimer's disease, the fibrils are mainly comprised of 40-42 residue fragments, AP(1-

40) and Ap(l-42), of the amyloid precursor protein (APP) (Heininger, 1999; McGeer and

McGeer, 1997; Sisodia and Price, 1995). The fibrils that are attributed to prion diseases such as CJD and BSE result from the aggregation of amyloidogenic scrapie prion protein

(PrPsc), which is the mutated isoform of the normal cellular prion protein, PrPc (Dandoy-

Dron et al, 1998; Sanada and Vo-Dinh, 2001; Slepoy et al, 2001). Under appropriate conditions, it appears that virtually all proteins can form fibrils, provided that they have the ability to adopt the P-sheet structure.

The devastating and often irreversible in vivo damage caused by these amyloidogenic proteins has been well documented and has created an immense challenge for the development of novel therapeutic solutions to the many diseases associated with

12 protein misfolding. For this reason, elucidation of the mechanisms governing the formation of amyloid fibrils has been very actively pursued. One approach has been to investigate the possible roles of molecular chaperones in the prevention of protein misfolding, since increases in their expression seem to be correlated with these and many

other diseases.

Figure 1.4: Images of amyloid fibrils. (A) Transmission electron microscopy of fibrils formed by the seven-residue peptide AM6-22, negatively stained with uranyl acetate [38]. (B) Cryo-electron microscopy model of an amyloid fibril derived from the SH3 domain of the p85a subunit of bovine phosphatidyl- inositol-3'-kinase; the stacking arrangement of the b structures is shown in colour (Jimenez et al, 1999).

1.2.4 Molecular Chaperones

The term molecular chaperones was applied by John Ellis to describe a diverse group of functionally related proteins that function in vivo to assist in protein assembly by specifically interacting with non-native protein structures and inhibiting their precipitation (Ellis, 1987; Ellis, 1992). It is believed that, during their chaperone action, molecular chaperones undergo structural changes that result in the exposure on their surface of hydrophobic regions. These regions allow the chaperones to interact with

13 hydrophobic domains of molten globule substrate proteins. In addition, it is believed that these interactions are promoted by highly flexible regions, which many chaperones possess.

Of the molecular chaperone superfamily, the best characterised chaperones belong to the stress-induced proteins that are collectively known as the heat shock proteins

(Hsps). As the name implies, Hsps first came to attention because of their increased expression in response to elevated temperature (heat shock). Numerous studies have now shown that, in additional to thermal stress, the increased expression of Hsps can also be induced by many other stresses (Mehlen et al, 1995; Preville et al, 1999). The five major heat shock protein families include Hsp 100/104, , , Hsp60/GroEL, and the small heat shock proteins (sHsps) (Kabakov and Gabai, 1997). While HsplOO,

Hsp90, and the sHsps (Gething, 1997; Jakob and Buchner, 1994; Sanchez and Lindguist,

1990) are expressed primarily in response to heat shock, Hsp70 and Hsp60 are involved in protein folding and refolding under normal conditions as well as under heat shock conditions (Gething, 1997; Haiti, 1996). Some of these heat shock protein families, notably the sHsps, HSP70 and HSP60/GroEL, will be discussed in more detail in subsequent chapters.

Many chaperones have been implicated in pathological states including those mentioned previously (see section 1.2.5; "Protein Misfolding and its Consequences").

Although their function in diseases associated with protein misfolding is not fully understood, their elevated expression in response to these diseases nevertheless raises the question as to what role chaperone proteins have in disease and whether their presence is beneficial (slowing the progression of disease) or detrimental (exacerbating disease).

Until recently, all known molecular chaperones were intracellular proteins. At present, only a small number of extracellular proteins with chaperone-like activity have been

14 discovered (Bhattacharyya and Das, 1999; Pavlicek and Ettrich, 1999). Of these, a near- ubiquitous protein called clusterin became the first reported extracellular molecular chaperone (Humphreys et al, 1999). For the remainder of this chapter, the general characteristics of clusterin and its predicted functions in vivo will become the main focus of discussion.

1.3 CLUSTERIN

The name "clusterin" was internationally accepted at the first Clusterin Workshop in 1992 at Cambridge, UK, to describe a protein which has the ability to promote the aggregation or "clustering" of a variety of cell types, including erythrocytes, renal epithelial cells, Sertoli and mouse testes TM-4 cells (Figure 1.5) (Blaschuk et al, 1983;

Rosenberg and Silkensen, 1995). This protein was initially isolated in 1983 from ram rete testis fluid by conventional methods and by immunoaffinity chromatography

(Blaschuk et al, 1983). Previously, many homologues of clusterin were isolated from different animal species and with each new discovery a different name was assigned to the protein (Table 1.2). For example, the human homologue of clusterin, first isolated in

1989, was subsequently named apolipoprotein J (apoJ), NA1/NA2 (de Silva et al, 1990a and b; Matsubara et al, 1995), SP-40,40 (Kirszbuam et al, 1989; Murphy et al, 1988;

O'Bryan et al, 1990), complement cytolysis inhibitor (CLI) (Jenne and Tschopp, 1989), and PADHC-9 (Rosenberg et al, 1993), depending on the locale in the human body in which it was discovered. Similarly, the rat homologue of clusterin was given the names

SGP-2 and TRPM-2 (Jenne and Tschopp, 1989; Law and Griswold, 1994; Wong et al,

1993). However, to avoid confusion, in this thesis the protein will simply be referred to as clusterin.

15 1" -»=•—»n— • * * it * f •»•m . » % I £.* » ? . t .* t ^ 1 % k L* 'i-i m . vf* 4 £ •' a j- , - •' * **4 % : ! , * *»^ ' .'» ; V IT 4 :^f A*

Figure 1.5: Microscopy images showing the "clustering" effects of clusterin on TM-4 cells. (A) In the absence of clusterin, TM-4 cells do not aggregate. (B) In the presence of exogenous clusterin, the cells aggregate extensively (Fritz et al, 1983).

1.3.1 Clusterin Structure and Biosynthesis

The mature clusterin molecule is a secreted 75-80 kDa heterodimer consisting of two non-identical 35-40 kDa subunits, a and P, that are joined together via a highly conserved motif containing five disulfide bonds (Choi-Miura et al, 1992). A schematic representation of this structure is shown in Figure 1.6. Southern blot analysis and cross- hybridisation studies have revealed that a single 1.6-2.0 kb mRNA, transcribed from a single copy gene located on 8p 12-21 in humans (Fink et al, 1993); chromosome 14 in mice (Jordan-Starck et al, 1994); and chromosome 15 in rats

(Goldner-Sauve et al, 1991), is responsible for encoding the polypeptide precursor of clusterin (Rosenberg and Silkensen, 1995; Rosenberg et al, 1993; Wilson and

Easterbrook-smith, 1996). The precursor is synthesised in tandem with a typical 21-22 amino acid , which directs the clusterin polypeptide to the lumen of the endoplasmic reticulum where the signal peptide is proteolytically removed. The disulfide bonds that hold the two subunits together are also formed at this point (Burkey et al,

1991).

16 o ON ON 00 ON ON ON OmN 00 OenN m NO ON ON ON 8 ON 00 ON D. ON ON oo 1 u ON 00 00 ON ON ON D. •c ON ON J= a ~- a O ' ' . C3 C3 a z u a *-1). <») •a •4} E •a r. ^ c cd 2

1) 60 cd o rN •m-t < OH § a a o 1—1 c^ e tt ,£5 ^ CN a a) O k 3 4—1 a .5 •s O 00 OH w 'Si u a on R: 5 v- o a ^ CJ s—•* a. ft o o .g. a o D. 1 X3 c a o .l-5H >N C2J4 .S x: < 4, O O Si o E GO CJ i p CJ -I S o o. 3 & < >N 3 H 04 a CJ 4+ < u o b H c^ U oo OH 5

17 To produce the mature clusterin molecule, which consists of 427 amino acid residues, the polypeptide is further cleaved between residues 205 (arginine) and 206

(serine), resulting in formation of the two subunits. The amino-terminal 205 amino acids form the a subunit whereas the carboxyl-terminal 222 amino acids constitute the p subunit. At this stage, the post-translationally processed product now has a molecular mass of 51 kDa (de Silva, 1990b; Jordan-Starck et al, 1994; Wilson and Easterbrook-

Smith, 1996), which is 25-29 kDa smaller than the mature clusterin molecule. Chemical deglycosylation of the clusterin-oc and -ft subunits by de Silva et al. (de Silva et al,

1990c) has revealed that this difference is due solely to extensive N-linked glycosylation of the mature protein. Rat Sertoli cell-derived clusterin, for example, contains 23-

30% (w/w) highly sulfated carbohydrates whereas carbohydrate accounts for 30% of the molecular mass of human clusterin (Griswold et al, 1986; Law and Griswold,

1994). Although it is generally believed that glycosylation helps to solubilise proteins in an aqueous environment, the absence of carbohydrates has no effect on the post- translational modification and secretion of clusterin. This clearly indicates that carbohydrates found on the surface of clusterin are not essential for proper clusterin trafficking and secretion (Wilson and Easterbrook-Smith, 1996). Yet, this suggestion seems to be contradicted by a recent report which showed that overexpression of an intracellular lectin called VIP36 stimulated the secretion of a number of secretory , including clusterin, by interacting with and facilitating the ER-to-plasma membrane transport of proteins with high mannose- type glycans (Hara-Kuge et al, 2002). Obviously, more research will be needed to determine the exact role of the carbohydrates found on the surface of clusterin.

Once glycosylated, the mature protein is usually secreted into the extracellular space. However, apparent exceptions include chicken clusterin which is retained as an

18 uncleaved single chain polypeptide (Mahon et al, 1999) and a truncated form of clusterin which, following the treatment of CCL64 and HepG2 cells with transforming growth factor beta (TGF0) (Reddy et al, 1996), was reported to accumulate in the nucleus of the cells.

The primary sequence of clusterin is known, but the secondary, tertiary and quaternary structures are yet to be determined. What is known, however, is that 40% of the total number of amino acids are polar while 41% are non-polar (Blaschuk et al, 1983). Furthermore, circular dichroism spectroscopy indicates that cc-helices account for 41% of clusterin structure, whereas P-sheets and turns account for approximately 19% and 32%, respectively. Structure predictions also suggest that mature form of the clusterin molecule contains extensive amphipathic oc-helical regions that are formed by the 76 N-terminal amino acid residues, as well as regions comprised of residues 150-170, 215-240, 300-350, and 406-420 (Figure 1.6) (Jenne et al, 1991; Jenne and Tschopp, 1992). The hydrophobic nature of clusterin, which results in its aggregation with itself to form dimers and tetramers, may arise from these predicted amphipathic regions (Blaschuk et al, 1983).

The degree of identity in the amino acid sequence of clusterin varies between species and is strongly correlated with the "phylogenetic distances" between these species (Figure 1.7). For example, clusterin isolated from mice and rats (both of which belong to the same phylum) have been shown to share 91% sequence identity

(Wong et al, 1993) whereas human and rat clusterin share 77% sequence identity

(Matsubara et al, 1995). Human and canine clusterin share 78% whereas human and bovine clusterin show 72% homology (Jenne and Tschopp,

1992). Clusterin from two relatively unrelated species (human and quail) show the least homology in their amino acid sequence, with 46-49% homology (Jenne and

19 Tschopp, 1992). However, clusterin from all species contains one region (residues

87-150) which exhibits the greatest degree of conservation with 98% sequence identity, suggesting that this region may be of structural and/or functional

importance.

a subunit

Y 205 3-c

i i i 3-N 427 X 206 B subunit

KEY

Glycosylation sites Predicted coiled-coil cc-helices

Milrl I I I I I Disulfide bonds Predicted amphipathic cc-helices I I I II

Figure 1.6: Schematic representation of the structure of clusterin. The main features of the clusterin structure shown include glycosylation sites, the disulfide bond motif, and the predicted coiled-coil a-helices and amphipathic a-helices (see key). The amino acid positions at the N-terminus (N-) or C-terminus (-C) of the two subunits are indicated as numbers. This figure was modified from figure 1 of [77].

A search of current protein-sequence databases, such as the Swiss protein

database, revealed no substantial sequence homology between clusterin and other known

proteins. Yet, discrete regions of similarity have been reported between clusterin and a-

helical coiled-coil proteins such as myosins and desmins (Tsuruta et al, 1990), oc-helical

proteins such as apolipoprotein A-I (Apo A-I) (de Silva et al, 1990c), and the cysteine-

rich thrombospondin type-I regions in terminal complement components C7, C8, and C9

(Jenne and Tschopp et al, 1992; Tschopp et al., 1993).

20 a

Cj LU < ' ' ' O LLI HCQ. T.f SEOO a nc & s "oo i ->J _l Q i i z H ' > • 0 uj UJ ZCO "J • B I O I UJ yj u- I a ! *P UJ ' =-1F Z ' L) S • i <£.* • : '. i 1 "-"L ,'inio mo i i_ r-. UJ ' Wn x i^, ^UJS ^ i , a !->-= • = ^f DC ' O Q. ol • i O =• ' ' „! "- u. Of) H Uj Vaa^ a mo UJ ^T "-x O • « 5t <« •' i, i S SDC~ ! t : ! f i ' •is x UJ UJUJ a: u- —•"fi —i — ar1 n UJ o • • i UJ IT I I IIJ i • i i a i 1 UJU J ^00 UJ( DC 1^ I ^ S ' • 1 —> s 1 1 a i O i >a <• 1 1 UJ • o • 5£ —1 1 u_ -I I • ' 5 a i Q ' _l I ^ < > m I I UJQ 2 t => i i •£ O a Q x a •< • i . — • S i uj I ' I I • OO UJ i o 9 u- I ^ I ' I I at- mi 1 UJ g 0ox a H i v I ' I i ,' UJ rr B ' , i P' '. a. i I i 2,1,1 aIE ' o ' 1 ' aUJ UJ Q Q Q j. I I I I 1 ' <. I ' I I- zOO az u- I | ?7 . DC^ uj ; o _ I . . DC ' •§ UJ I'll I1 = UJ I I a' Uj I I • I _i • i *: zU) ZOTP i- ' 1 u iLiq n (0 • o s >-• i UJ ' 41 s:i- -_ UJ ••• »» :: =r Q [ ODC O i—i i . 1', < Z O -^ i DC • X • a • ' a • ,1 1 ,UJ O ' 1 1 U_ | ^ UJ ^ 2 i ' ', on > 1 ' J u- ^ • y_ I -:•.;? z • 5« UJ I ' o a • S i < 1 UJ (5 a i =. o 1U I T DC OO a g i > 1 >1 v_ I u 'X OHO Uj 3- i i i (0 1 u- ' ^X^-Jo uc i ; I— ' ' ™ I z o 1 1 0, • °-°-X- UJ ' ' •¥ = > ; «? a-—I • ' ' ^ o I- oo 3 I, I D. 1 = =. ; , i • < i 5 > «. , ,i 35: ->i a ' •l- x £ I 1.'° ', ! 1 DC I I I I I a ' ' oo << 0= ] i z • ' 1 1 i i 0i0 aI i 1 1 •82' ^oi .. ' CI -^ 1 ' I 1 U_ I 1 OJ , i m > ' I ' UJ • 1 1 f 9^ 1 1 ul-S G _3 O i 11 UJ • 1 1 ^ii;ix 1 1 1 1 1 : i ' < « r S 5 <.-!•?• < 1 1 -11 g< I u-> ?.' '• aDC •' = • li_ —I u-i a ' * * UJ ' 1 1 17 1 1 ' •' '. a. ZOTJ UJ & ° ^ - i i i > >o i' • UJ I I ' T „. i i on > > OJ 05 U j. < O i 1 1 Ift U QJ 1 1 IE ' • O • on £ I O'i UJ UJ ^ U-1 UJ O« i' 1 1 • 1 1 • I u- "- i DC ' 1 ' ji l a , UjUJ < B. i£| 1^^ 1 i z Hi- H * •< ' -^(0 00 . OO , i i • >Li- '1 X ' I > DC 5:97 9 i a J_ 1 >DC E O ' ' ' 1' 1 S i 1 1 H '.°0 >- • o O ' , Q ' I

21 1.3.2 Clusterin Distribution and Binding Interactions

Clusterin is regarded as a near-ubiquitous protein because it is found in virtually all mammalian tissues, where the levels of expression are very similar between different species (Wilson and Easterbrook-Smith, 1996). Strong levels of expression have been detected in the adrenal medulla, brain, anterior and posterior pituitary, liver, ovary, and testis, with low levels in the lung, lactating mammary tissue, , spleen, thymus, and the uterus (de Silva et al, 1990b; Duguid et al,

1989; Boggs et al, 1996; Collard and Griswold, 1987; Oda et al, 1994). Recent reports have also indicated the presence of clusterin in the corneal epithelium, retina, and aqueous humor of the eye (Dota et al, 1999). However, this is not to say that all cells in these organs express clusterin. For example, in the prostatic ductal system and the epididymis, only endothelial cells express clusterin mRNA. In addition, whereas atrial myocytes of the heart and Chief cells of the stomach both produce clusterin, ventricular myocytes (heart) and parietal cells (stomach) do not (Rosenberg et al, 1993). In many organs, clusterin is often expressed at fluid-tissue interfaces; this has led to the suggestion that it may serve to protect cell membranes from the harmful effects of prolonged exposure to "biologically active fluids" such as bile, gastric juice, pancreatic juice, gland and urine (Jordan-Starck et al, 1994;

Aronow etal, 1993).

Clusterin is also present in a number of body fluids including breast milk, cerebrospinal fluid, plasma, semen, and urine (Jenne and Tschopp, 1989; O'Bryan et al, 1990; Aronow et al, 1993). The concentration of clusterin in human plasma has been reported to range between 35 and 105 p.g/mL (0.44 - 1.35 uM) (Jenne and

Tschopp, 1989; Murphy et al, 1988). In human seminal fluid, the concentration of clusterin is approximately 10-fold higher (Matsubara et al, 1995) while in the

22 cerebrospinal fluid (CSF) clusterin concentration varies between 1.2 and 3.6 p.g/mL

(Calero et al, 2000). Serum clusterin is produced by megakaryocytes, which are also responsible for the synthesis of platelets in the bloodstream. Clusterin is not actually released from platelets until they degranulate in response to endothelial damage. The ability of clusterin to promote aggregation of a variety of cell types including erythrocytes and white blood cells have, in the past, led to suggestions that clusterin may be involved in cell development and (Silkensen et al,

1999).

Extensive work on mouse clusterin mRNA during various stages of gestation and organogenesis have shown that the location of clusterin expression or production may change at different stages of development. For example:

• Clusterin is present in both the atria and ventricles of the foetal mouse heart but only in the atria of the adult heart (Aronow et al, 1993);

• Clusterin is present in the bronchial tree of the developing human lung but is absent in the mature lung (Rosenberg and Silkensen, 1995), and;

• Clusterin expression increases in the central nervous system as embryogenesis progresses towards the postnatal stages of development (Rosenberg and Silkensen,

1995).

The stage-dependent expression of clusterin suggests possible roles in organogenesis and development.

Clusterin interacts with a variety of proteins and lipids in vitro. Some of the best known binding partners of clusterin include terminal complement components

C7, C8, and C9 (Boggs et al, 1996; Choi et al, 1989; McDonald and Nelsustuen,

1997), immunoglobulins (Wilson and Easterbrook-Smith, 1992), Alzheimer's

23 amyloid beta molecules (Ghiso et al, 1993; Zlokovic et al, 1994; Matsubara et al,

1995; Choi-Miura and Oda, 1996), and a serum protein called paraoxonase (Kelso et al, 1994). Other binding partners include glutathione S-transferase (GST)

(Hochgrebe et al, 2000), heparin (Pankhurst et al, 1998), myosins and desmins, and

SIC (Streptococcal inhibitor of complement-mediated lysis) (Akesson et al, 1996).

In human blood, clusterin is associated with high-density lipoprotein (HDL) particles along with another plasma protein called apolipoprotein A-I (apoA-I), which is often co-purified with clusterin (Jenne and Tschopp, 1989). Clusterin aggregates with itself under conditions of neutral pH and low salt concentrations to form dimers and tetramers (Blaschuk et al, 1983; Griswold et al, 1986). In addition to the hydrophobic nature of clusterin, a combination of non-ionic and weak electrostatic forces is believed to be responsible for these interactions. It is not known why clusterin interacts with so many biological molecules or why it binds with itself.

1.3.3 Proposed Biological Functions of Clusterin

For many years, the process of identifying a functional role for clusterin has been complicated by its near-ubiquitous distribution and its ability to bind to a variety of proteins and lipids (see above). Numerous biological functions have been proposed for clusterin, further complicating the quest to find its true function/s. Clusterin has been implicated in the biological processes of reproduction (Blaschuk et al, 1983; Law and

Griswold, 1994); Tenniswood et al, 1998), lipid transport (Jenne et al, 1991), endocrine secretion (Rosenberg et al, 1993), apoptosis or programmed cell death (Jenne and Tschopp, 1992; Wong et al, 1993; French et al, 1994; Wilson and Easterbrook-

Smith, 1996), and complement regulation (Choi et al, 1989; Kirszbuam et al, 1989;

McDonald and Nelsustuen, 1997). Rosenberg and Silkensen (1995) previously raised the

24 question of whether clusterin is truly a multifunctional protein or has a single unifying function that explains its distribution among the various physiological settings in which it has been found. Some aspects of proposed clusterin functions are discussed in detail below.

1.3.3.1 Clusterin and reproduction

Questions as to the functions of clusterin within the reproductive system were raised after the initial isolation of this major glycoprotein from ram rete testis fluid.

Clusterin constitutes about 18% of the total protein secreted into the testis fluid and has the ability to elicit the clustering of Sertoli cells (Blaschuk et al, 1983). Sertoli cells provide an appropriate environment for spermatogenesis and the transfer of nutrients to the developing spermatocytes and spermatids. They are also the major testicular source of the rat homologue of clusterin, SGP-2, which comprises 50% of all protein secreted by these cells (Law and Griswold, 1994; Rosenberg and Silkensen, 1995).

Clusterin is present in the male reproductive tract in the cytoplasm of various cell types, including Sertoli cells and epididymal cells, the late developing spermatids and released spermatozoa in the testis, and on the surface of maturing sperm cells in the epididymis (Collard and Griswold, 1987; O'Bryan et al, 1990; Aronow et al, 1993;

Rosenberg and Silkensen, 1995). Due to its high abundance and broad distribution, it has been suggested that clusterin plays a major role in the generation and maturation of sperm cells, especially the formation of the cytoskeletal components of sperm tails (Law and

Griswold, 1994). However, this remains to be substantiated. Other suggested functions of clusterin in the male reproductive tract include lipid transport, mediation of cell-cell interactions (e.g. of Sertoli cells) and the protection of spermatozoa from harmful and degradative environments (Rosenberg and Silkensen, 1995). In addition, the localisation

25 of clusterin on male germ cells raises the possibility that it may protect sperm from complement lysis in the vagina post-coitus. In females, clusterin mRNA and protein have been found in the ovary and uterus (Ahuja et al, 1994). Clusterin has been detected in granulosa cells during the development of the ovarian follicle where it has been postulated to protect and support the maturation of the ovum.

1.3.3.2 Involvement of clusterin in lipid transport

The highly amphipathic nature of clusterin allows it to bind molecules with similar properties - hence, the existence of numerous binding partners (see section 1.3.2).

As Jenne et al. reported (Jenne et al, 1991), clusterin purified by monoclonal antibody- affinity chromatography was often isolated in association with a common 28 kDa protein of human plasma. Partial amino-terminal sequencing revealed the identity of this protein to be apoA-I. It is now known that the majority of serum clusterin does not circulate as a free globular molecule but rather is associated with high-density lipoprotein (HDL) particles in the form of clusterin-HDL-apoA-I complexes. These complexes are rich in free cholesterol but poor in lipids, with lipid composition ranging from as low as 11% (de

Silva et al, 1990c), to a moderate 22% (Jenne et al, 1991). Since clusterin is commonly associated with apoA-I, it has been suggested that clusterin is involved in lipid transport

(Rosenberg and Silkensen, 1995), especially the transport of cholesterol from peripheral tissues to the liver (de Silva et al, 1990c). ApoA-I is an apolipoprotein, the protein components of micellar complexes of lipoproteins that are responsible for transporting otherwise insoluble lecithin, triglyceride and cholesterol in vertebrate blood (Stryer,

1995). Supporting the proposal that clusterin is involved with lipid transport is the observation that levels of cholesterol and clusterin in human plasma are positively correlated and that clusterin is found at high levels in atherosclerotic plaques (Jordan-

26 Starck et al, 1994). A role for clusterin in the uptake and redistribution of lipids from damaged cells has been proposed which is consistent with evidence showing that clusterin-HDL-apoAI complexes have an affinity for membranes of damaged, abnormal, or dying cells (Rosenberg and Silkensen, 1995). Clusterin is a ligand for the low- density lipoprotein receptor-related protein-2/megalin/gp330 (LRP-2), an endocytic receptor of the low density lipoprotein receptor family (Kounnas et al, 1995;

Morales et al, 1996; Hermo et al, 1999). In vivo, LRP-2 has been shown to mediate the endocytosis of clusterin alone, or clusterin associated with cholesterol-rich lipoproteins. Since LRP-2 is also believed to play a crucial role in the development of epithelial cells of the efferent ducts and epididymis as well as in the uptake and disposal of lipoproteins and cholesterol, it has been suggested that clusterin may act as a marker for the disposal of its lipid substrates (Hermo et al, 1999). Human clusterin has been reported to promote the efflux of cholesterol from macrophage foam cells of apoE-null mice, suggesting that clusterin may be involved in cellular cholesterol homeostasis (Gelissen et al, 1998). Furthermore, this report suggested that, by being able to regulate cholesterol transport, clusterin might have anti­ atherogenic properties. However, this remains to be established.

1.3.3.3 Clusterin and apoptosis

There are two different ways by which mammalian cells die in vivo (Figure 1.8).

Cell death may be due to necrosis, or accidental cell death, which is a relatively uncontrolled process usually caused by traumatic physical injuries or chemical changes to the cellular environment (French et al, 1994). Alternatively, cells that are no longer needed by an organism may die as a result of apoptosis or programmed cell death (PCD), which is characterised by distinct morphological and biochemical changes. These

27 morphological changes include a marked decrease in cell volume (cell shrinkage), cytoskeletal modification resulting in membrane blebbing, nuclear condensation and fragmentation, degradation of DNA into oligonucleosomal fragments, as well as the formation of "apoptotic bodies" (French et al, 1994). In vivo, apoptotic bodies are always rapidly phagocytosed by macrophages to ensure that any proteolytic and other lytic enzymes are not released into the extracellular environment, thereby preventing localised inflammation (Kuby, 1997).

The association between clusterin and apoptosis came about due to reports showing an elevated expression of clusterin specifically in apoptotic cells. Some of these reports suggested that clusterin is an apoptotic marker (Buttyan et al, 1989; Danik et al,

1991). Clusterin's apparent importance in apoptosis has been highlighted by findings that in certain diseases, clusterin mRNA is usually upregulated. For instance, in 1989,

Duguid et al reported an increase in clusterin expression in the hippocampus of patients suffering from Alzheimer's or Pick's disease (Duguid et al. 1989). In the same year,

Bettuzzi et al. reported up-regulation of TRPM-2 (rat clusterin) in involuting prostatic epithelial cells undergoing apoptosis following chemical or surgical castration (Bettuzzi et al, 1989). Since then, elevated TRPM-2 expression has also been reported in regressing mammalian glands after weaning, in the renal collecting ducts and distal tubules after ureteral obstruction, and in a variety of carcinomatous cells undergoing cell death following the removal of trophic hormones which normally sustain their survival

(Wong et al, 1993). More recently, increased clusterin expression has been observed in populations of photoreceptors undergoing apoptosis in retinal degeneration slow (rds) mutant mouse retinas (Agawal et al, 1996) and light-induced degenerative rat retinas

(Jomaryeffl/., 1999).

28 NECROSIS APOPTOSIS

Mild convolution Chromatin compaction and segregation Chromatin clumping Condensation of Swollen organelles cytoplasm Flocculent mitochondria I i

Nuclear fragmentation • ^J Membrane blebbing %J*G} Apoptotic bodies

Disintegration I Phagocytosis Release of Apoptotic body intracellular contents

I Phagocytic cell Inflammation (macrophage)

Figure 1.8 : Schematic outline of the comparative morphological differences that occur in necrotic and apoptotic cells. Necrosis or accidental cell death is characterised by the loss of plasma membrane integrity and cell rupture, which causes the release of intracellular contents and localised inflammation. Apoptosis or 'programmed' (non-pathological) cell death involves cytoplasmic and nuclear condensation, membrane blebbing and formation of apoptotic bodies, followed by phagocytosis. In contrast to necrosis, apoptosis does not result in inflammation. Source: Kuby, 1997.

Numerous hypotheses regarding the involvement of clusterin in apoptosis have been proposed. It has been suggested that clusterin may directly activate an apoptotic suicide mechanism, enhance the opsonization of cell debris, or even function as a chemoattractant for phagocytic cells by 'labelling' or marking cell debris for removal

29 (Wong et al, 1993; French et al, 1994). Initial reports suggested that cells over- expressing clusterin are themselves apoptotic. However, it is important to note that this is not always the case. For instance, cells of the skin, intestinal epithelium, and placental decidual cells that are constantly undergoing apoptosis do not seem to express clusterin at levels that can be measured (Aronow et al, 1993). In addition, in vitro studies using human thymus cells have shown that clusterin gene expression occurs much more prominently in non-apoptotic cells that are in the immediate vicinity of cells undergoing

PCD (French et al, 1994). Therefore, it is reasonable to suggest that clusterin expression may not be an initiating factor in apoptosis, but rather appears as a consequence of it.

Indeed, clusterin has been shown to protect a wide range of cells from apoptotic cell death mediated by various stimuli that include amyloid-p" (1-40) neurotoxicity (Boggs et al, 1996), oedematous caerulein-induced pancreatitis (Calvo et al, 1998), ultraviolet B irradiation (French et al, 1994), hypoxia-induced ischemia (Hee Han et al, 2001), TNF- oc-mediated cytotoxicity (Humphreys et al, 1997; Sensibar et al, 1995), peroxide

(H2O2)- induced oxidation (Schwochau et al, 1998), heat shock (Viard et al, 1999) and the withdrawal of growth factors such as pregnant mare serum gonadotropin (PMSG)

(Zwain and Amato, 2000).

The correlation between elevated levels of clusterin mRNA and the survival of carcinoma cells such as human renal clear cell carcinoma (Parczyk et al, 1994) and

RUCA-I rat endometrial adenocarcinoma (Bora et al, 1993) also led to the suggestion that clusterin may promote cell survival by suppressing the terminal complement cascade.

If this is true, and if clusterin can truly inhibit complement-mediated lysis in vivo (see section 1.3.3.4), its expression during development might prove to be a significant hindrance to possible cancer therapies that sought to utilise the body's own to target and dispose of the rogue cells.

30 1.3.3.4 Clusterin in complement regulation

As a part of the general vertebrate immune system, the complement system consists of at least 29 different serum and membrane proteins (C1-C9, factors B and D, and a series of regulatory proteins) and is responsible for causing a variety of immune responses, such as localised inflammation and lysis of foreign microorganisms (reviewed in [Kuby, 1997; Holers et al, 1985; Liszewski et al, 1991). The activation of complement is triggered either by bound antibodies (to initiate the classical pathway) or polysaccharides (the alternative and lectin pathways) on the surface of foreign cells and often causes the dilation of blood vessels and accumulation of phagocytes at sites of infection. One distinct feature of the complement system is the assembly of complement proteins (C5b, C6, C7, C8, and C9) into a membrane attack complex (MAC), which, upon insertion into the cell membrane, is responsible for cell lysis resulting in cell death

(Kuby, 1997). The formation of the MAC is highly regulated and if any one of the complement components fails to associate correctly or bind to the target membrane, all lysis will not eventuate (Kirszbuam et al, 1989).

The demonstrated ability of clusterin to bind with high affinity to purified terminal complement proteins C5b, C7, C8 and C9 led to the proposal that it functions as a complement regulator (Murphy et al, 1988; Tschopp et al, 1993). In numerous in vitro studies, the binding of clusterin to complement proteins was shown to prevent formation of the MAC, resulting in the inhibition of complement-mediated cytolysis of a variety of cell types. Choi-Miura et al, 1989, reported that the complement-dependent hemolysis of erythrocytes could be inhibited by the incorporation of clusterin prior to the addition of C7 to C5b6. However, when clusterin was added simultaneously with C5b6,

C7, C8 and C9, complement-mediated cytolysis was enhanced. This observation was interpreted as evidence that clusterin inhibits complement by masking the membrane-

31 binding site of C5b-7 thus preventing its association with the membrane, and hence, cell lysis. The association of clusterin with C5b-7 may be of significant importance in the prevention of complement-mediated lysis of healthy bystander cells by fluid phase C5b-7 complexes that have missed their intended targets. In situ experiments conducted by

McDonald and Nelsestuan in 1997 on rat kidneys supported this hypothesis by showing that complement-induced glomerular injury could be augmented in kidneys that are perfused with clusterin-depleted plasma. Similarly, glomerular damage in the rat kidney upon complement activation was enhanced when perfused with human plasma depleted of clusterin to 30% of the initial concentration (Saunders et al, 1994). Taken together, the above-mentioned reports suggest that clusterin may function as a complement regulator to inhibit complement-mediated cytolysis.

In contrast, a recent report challenges the idea that clusterin functions as a complement regulator by presenting data showing that, at physiological plasma concentrations (i.e. reported as ranging from 50 - 150 p.g/ml), exogenous clusterin could not protect (i) sheep or rabbit erythrocytes against classical or alternative complement pathway mediated lysis, (ii) human erythrocytes undergoing insulin-induced complement mediated lysis, and (iii) transfected L929 cells from complement-mediated lysis in human serum (in this model, clusterin was expressed as a cell surface protein) (Vakeva et al,

1993). In the case of undiluted serum, results showed that approximately 30 mg/ml clusterin was required to protect 50% of sheep or rabbit erythrocytes against lysis mediated by undiluted serum, which equated to a clusterin concentration at least two orders of magnitude greater than that in normal human serum. As a result of these findings, it was concluded that clusterin might not be a physiologically relevant regulator of complement activation. This supported an earlier finding that clusterin colocalised with the MAC on the surface of damaged cardiomyocytes but was not present on the

32 surface of undamaged cardiomyocytes, and that it was unable to protect myocardium against complement-mediated attack (Hatters et al, 2002). Therefore, a genuine role of clusterin as a physiological regulator of complement remains to be established.

1.3.3.5 Clusterin as a molecular chaperone

A common theme reflected in the many functions proposed for clusterin is that it is implicated as a component of the general response to cellular and tissue injuries.

Clusterin has been associated with a variety of disease states including Alzheimer's disease (May and Finch, 1992; Choi-Miura and Oda, 1996), scrapie (Duguid et al,

1989) renal tubular injury (Rosenberg et al, 1993), (Fink et al,

1993), epilepsy (Danik et al, 1991), and retinal degeneration (Jomary et al, 1999), just to name a few. The role played by clusterin in these diseases is still unknown.

However, recent findings suggest that clusterin may be involved in protein unfolding in vivo and that it may function as a molecular chaperone in the diseases with which it is associated. Many other chaperones, such as the Hsps, are also associated with the same disease states as clusterin (section 1.2.6).

As mentioned earlier, in vitro studies conducted by Humphreys et al, 1999, showed, for the first time, that clusterin has functional similarities to the small heat shock proteins, a group of intracellular molecular chaperones which function in vivo to protect other proteins from stress-induced unfolding, aggregation and precipitation. This report highlighted clusterin's ability to interact with, and stabilise, numerous unrelated stress- affected target proteins to form soluble high molecular weight (HMW) complexes. These substrates included heat-stressed catalase and glutathione S-transferase (GST), and chemically reduced a-lactalbumin (a-lac) and bovine serum albumin (BSA). Enzyme- linked immunosorbent assays (ELISAs) revealed that, like the sHsps, clusterin binds

33 preferentially to stressed proteins. Although clusterin was able to inhibit stress-induced precipitation of test proteins, it was unable to protect catalase and GST against stress- induced loss of activity.

Recently, Hatters et al demonstrated that clusterin, at substoichiometric ratios (1 :

30-100, clusterin:apoC-H) could suppress in vitro amyloid formation by apoC-II, a lipoprotein lipase activator, but failed to promote fibril dissociation (Hatters et al, 2002).

This behaviour of clusterin was consistent with the report by McHattie and Edington, who demonstrated that the in vitro aggregation of the neuropeptide PrP106-126 to form amyloidal fibrillar structures could be prevented in a dose-dependent manner by clusterin

(Mc Hattie and Edington, 1999). More specifically, 100 u.g/ml of clusterin was able to suppress the aggregation of PrP106-126 by as much as 90% with only a further 5% reduction in aggregation in the presence of 1 mg/ml clusterin.

The manner in which clusterin interacts with its substrates is still under investigation. One suggestion involves the binding of clusterin to its substrates through hydrophobic interactions, in much the same way as Hsp-substrate binding is believed to occur. By sequence analysis and ANS binding, clusterin has recently been shown to contain three flexible, molten-globule-like regions that are highly conserved among the different mammalian clusterin homologues (Bailey et al, 2001). A prominent feature of these regions of predicted disorder, approximated to the N- and C-termini of the a- and

|3-subunits, respectively, and the N-terminal region of the ^-subunit, is the presence of amphipathic a-helices. It is the hydrophobic surfaces of these a-helices that are believed to be important for clusterin binding to the cell membrane of sperm (Law and Griswold,

1994) and to the various, naturally found or stress-affected ligands, most of which are predominantly hydrophobic.

34 1.4 THESIS OBJECTIVES

The report by Humphreys et al, 1999 highlighted clusterin's ability to stabilise a wide range of unrelated stress-affected proteins and provided the first step towards an understanding of the possible role of clusterin in the processes of protein unfolding and refolding. This project aimed to further characterise the structure-function relationship of clusterin and to address a number of uncertainties relating to its role as a molecular chaperone. The following questions were addressed:

1) Can the apparent generality of clusterin's chaperone activity be extended further to include other, previously untested, substrate proteins such as ovotransferrin (also known as conalbumin), lysozyme, and y-crystallin? This was to firmly establish that clusterin is a molecular chaperone with broad substrate specificity.

2) What effects do temperature andpH have on the chaperone activity of clusterin? The action of certain Hsps, such as a-crystallin, is enhanced at elevated temperature; the effect of pH is unknown.

3) Can clusterin preserve the enzymatic activity of stressed catalase and alcohol dehydrogenase? Catalase was chosen to confirm previous findings by David Humphreys, who had shown that clusterin was unable to prevent heat-induced loss of catalase activity.

Alcohol dehydrogenase, on the other hand, had not been tested.

4) Does clusterin possess ATPase activity? If clusterin had ATPase activity it might indicate that clusterin could refold its substrates, since most chaperones that are stoichiometrically able to refold proteins require ATP to do so.

5) If clusterin is incapable of refolding its substrates, can it stabilise stressed proteins in a state competent for refolding by other chaperones (eg. Hsc 70)?

6) Can recombinant DNA technology be used to identify the structural domains responsible for the function of clusterin as a molecular chaperone?

35 Chapter 2

General Materials and Methods

36 2.1 GENERAL BUFFERS AND SOLUTIONS

Buffers that were used throughout this project include: phosphate buffered saline

(PBS; 137 mM NaCl, 2.7 mM KC1, 1.47 mM KH2P04, 11.6 mM Na2HP04, pH 7.4),

sodium phosphate buffer (50 mM Na2HP04, pH 7.0), and lx heat-denatured casein

(HDC; 1% (w/v) casein, 0.01% (w/v) thimerosal, in PBS, pH7.4). Other less commonly used buffers and solutions can be found in their respective sections. A list of chemicals used in this project, along with their full names, abbreviations, corresponding formulae and their suppliers, has been placed at the end of this chapter (Table 2.1). The suppliers of general consumables that are not listed in Table 2.1 are shown in brackets.

2.2 SDS-POLYACRYLAMIDE GEL ELECTROPHORESIS (SDS-PAGE)

2.2.1 SDS-PAGE Buffers and Stains

The compositions of the various buffers and stains used during SDS-PAGE are shown in Table 2.1. The resolving and stacking gels for Tris-glycine SDS-PAGE were prepared using the recipes outlined in Sambrook et al, 1989.

Reagents Composition

2x Sample buffer 0.005% (w/v) bromophenol blue, 20% (v/v) glycerol, 5% (w/v) SDS, 0.5 M Tris HC1, pH 8.8.

lx Running buffer 0.1% (w/v) SDS, 0.25 M glycine, 0.025 M Tris buffer, pH 8.3.

Coomassie blue stain 0.25% (w/v) coomassie brilliant blue (R250), 100 mL glacial acetic acid, 450 mL Methanol, 450 mL RO water.

Destain solution 100 mL glacial acetic acid, 450 mL Methanol, 450 mL RO water.

Table 2.1: Compositions of the buffers and stains used during SDS-PAGE.

37 2.2.2 SDS-PAGE Methods

2.2.2.1 Loading Protein Samples and Performing Electrophoresis - Samples containing a minimum of 10 \ig of protein were usually prepared in PBS to which an equal amount

(v/v) of lx sample buffer (Table 2.1) was added. For reduced samples, a 1% final concentration of 2-mercaptoethanol was also added. Samples were boiled for 5 minutes before being loaded into the wells. Running buffer (Table 2.1) was subsequently poured into the designated buffer chambers before the gel was electrophoresed at a constant 130

V until the dye front had reached the bottom of the gel. SDS-PAGE was performed using a Novex® "XCELL E" SDS-PAGE unit connected to a Bio-Rad model 1000/500 power unit.

2.2.2.2 Staining and Destaining - Once the electrophoresis had been completed, the gel was placed into a Coomassie Blue staining solution (Table 2.1) and left to stain for approximately 1 to 1.5 hours. The gel was then destained with 'destain' solution (Table

2.1) until it had been sufficiently destained and the bands were clearly visible.

2.2.2.3 Gel Drying and Storage - The gel drying procedure was performed in accordance with the manufacturer's specification (Novex, Sydney). Basically, this procedure involved he destained gel being placed into a Gel-Dry™ drying solution for 20 minutes before drying overnight at room temperature, sandwiched between two sheets of cellophane, in a Novex gel-drying frame. Once dried, the cellophane-sandwiched gel was lifted from the frame and taped on all sides to prevent the two sheets of cellophane from separating. The gel was then stored flat within a book.

38 2.3 AGAROSE GEL ELECTROPHORESIS

Agarose gel electrophoresis was used for the electrophoretic separation of native proteins as well as for DNA samples. The techniques used for electrophoretic separation of these two types of samples were basically identical, with the exception of the buffer components used.

2.3.1 Buffers for Native and DNA Agarose Gels

The compositions of the various buffers and stains used during native and DNA agarose gel electrophoresis are shown in Tables 2.2 and 2.3, respectively.

Reagents Composition

lx Running buffer 40 mM Tris, 1.142 mL glacial acetic acid, 1 mM EDTA, pH 7.4.

Sample Buffer 1% (w/v) Bromophenol blue, 10% (v/v) glycerol

Staining solution Colloidal coomassie G-250 (Noveline, Australia)

Destain solution 100 mL Glacial acetic acid, 450 mL methanol, 450 mL RO water.

Table 2.2: Compositions of the various reagents used during native agarose gel electrophoresis.

Reagents Composition

lx Running buffer 40 mM Tris, 1.142 mL glacial acetic acid, 1 mM EDTA, pH 7.4.

6x Sample Buffer 1% (w/v) Bromophenol blue, 10% (v/v) glycerol

Table 2.2: Compositions of the various reagents used during DNA agarose gel electrophoresis.

39 2.3.2 Agarose Gel Electrophoresis Set-up

The agarose gel electrophoresis unit used was the "Mini-Sub™ DNA Cell" system (Bio-Rad Laboratories, Sydney). A Bio-Rad model 1000/500 power pack was connected to this electrophoresis unit.

2.3.3 Agarose Gel Electrophoresis Methods:

2.3.3.1 Casting and running agarose gels - 1% agarose gels were prepared by heating and dissolving 1 g of analytical grade agarose (Promega, Sydney) in 100 mL of lx running buffer (or at an equivalent final concentration) in a microwave oven. For DNA gels, ethidium bromide (EtBr) was added to the dissolved agarose solution (after it has been allowed to cool to approximately 40 - 50 °C) to a final concentration of (0.5 pg/ml). Once dissolved, the gel was poured into an appropriately-sized gel casting tray before a comb was inserted to form the wells, and the gel was allowed to set at room temperature. Once the gel had set, the comb was removed before samples, containing a minimum of 10 pg protein or DNA in the appropriate sample buffer, were loaded into the wells. Thereafter, running buffer was applied to the buffer chambers located at both ends of the gel and electrophoresed at a constant 60 V for approximately 2-2.5 hours or until the dye front had reached the end of the gel.

2.3.3.2 Visualisation of DNA or Protein Bands - DNA, stained with EtBr and separated on the basis of size by agarose gel electrophoresis, was detected as bright orange bands when viewed under ultraviolet light. Gel images were captured by a digital video camera, which was mounted above the viewing platform on which the gel was placed to be viewed. Native proteins were visualised only after they had been transferred onto a membrane (PVDF or nitrocellulose), where they were stained with Colloidal coomassie

G-250 stain (see below).

40 2.4 WESTERN (PROTEIN) TRANSFER

Western transfer is a procedure by which proteins are transferred onto a membrane (e.g. Polyvinylidene difluoride (PVDF; Brasatec, South Australia) or nitrocellulose membranes). Once transferred, these proteins can either be stained or, more commonly, probed with, for example, enzyme-labelled antibodies; refer to section 2.5). Proteins can be transferred electrophoretically from SDS-PAGE gels or by capillary action from native agarose gels. The general set-ups for these two western transfer procedures are shown in Figure 2.1 and 2.2. In this project, membrane-bound proteins (such as clusterin) were stained with Colloidal Coomassie G-250 (Table 2.2) for approximately 5 minutes, followed by a longer period of destaining with destain solution (Table 2.2). Thereafter, the membrane was immuno-labelled and developed with enhanced chemiluminescence detection (ECL).

2.5 IMMUNO-LABELLING OF MEMBRANE-BOUND PROTEINS

The first step in the labelling of membrane bound proteins involved the blocking of potential binding sites on the membrane to minimize non-specific binding of immunological reagents. This was achieved by incubating the membrane in heat- denatured casein (HDC; section 2.1) for 1 hour at room temperature, after which, the

HDC was removed with several washes using lx PBS. This was followed by the addition of the primary antibody (at 10 pg/ml in HDC; raised in mice and is only specific for the protein of interest) and left at room temperature for another hour. After rinsing the membrane in lx PBS to remove excess, unbound antibodies, a solution containing the secondary antibody (sheep anti-mouse Ig-HRP (horseradish peroxidase) conjugated antibody (Silenus, Melbourne, Australia) at 1:1000 dilution in HDC) was added and

41 left at room temperature for one hour after which, the membrane was washed and developed with ECL.

Clamp Sequence for applying transfer materials

(applied to the white side)

1. Blotting sponge #1 2. One sheet of blotting paper White ooooo side 3. Single membrane (PVDF or Nitrocellulose) ooooo 4. Gel 5. One sheet of blotting paper ooooo 6. Blotting sponge #2 Figure 2.1a : Schematic representation o o of the 'mini gel holder cassette' for the Black Bio-Rad western transfer unit. Shown side on the right of it is the sequence in which transfer materials were applied onto the cassette. (Source: This diagram was drawn from observation)

Electrical leads to power pack

Mini Gel Holder Cassette (Black side)

Figure 2.1b : A schematic (Black side) representation of the general setup for the Bio-Rad western transfer unit for transferring proteins from SDS-PAGE gels onto a membrane. (Source: This diagram was Stirling rod drawn from observation) Weight

1 Stack of absorbent paper

4 Sheets of blotting paper

Direction of buffer uptake by capillary action Nitrocellulose membrane

Agarose gel

2 Sheets of blotting paper with ends folded to facilitate buffer uptake 1x Native Gel Electrophoresis Running Buffer

Glass Plate

Figure 2.2: Protein (Western) transfer by capillary action. The passive transfer of proteins from a 'native' agarose gel onto a nitrocellulose membrane was achieved by capillary transfer of buffer from a central reservoir (tray), through the gel, membrane, blotting paper, and into a stack of absorbant paper situated at the top of the set-up. This procedure usually required 12-16 hours to complete. The stack of highly absorbant paper was simply toilet paper that had been cut to size.

(Source: This diagram was drawn from observation)

43 2.6 ENHANCED CHEMILUMINESCENCE DETECTION(ECL)

ECL is a procedure (performed in the dark room) in which membrane-bound and immuno-labelled proteins are detected on photographic film. This procedure has been summarised as follow:

A detection solution comprising of equal volumes of the 'stable peroxide' and

'luminol/ enhancer' solutions (collectively known as the enhanced chemiluminescence or

SuperSignal® substrate working solution; for detection of HRP labels; Pierce, Sydney,

Australia) was spread over a membrane and left to develop for 1-2 min. For the detection of picogram (10"12 g) or greater amounts of HRP-lebelled proteins,

SuperSignal® WestPico was used. For the detection of femtogram (10"15 g) amounts of

HRP-lebelled proteins, SuperSignal® WestFemto (Pierce, Sydney) was used. The membrane was then drained of the detection solution, wrapped up in plastic food wrap and placed, with a small sheet of autoradiography film (Du Pont) placed on top, into a light-proof cassette containing a Kodak 'X-Omatic Intensifying Screen' and allowed to develop for a short period of time (ranging from a few seconds to a few minutes depending on the level of chemiluminescence displayed by the membrane-bound proteins). Once developed, the film was taken out of the light-proof cassette and placed into a solution of 'developer' until the bands became visible. The film was then transferred into a 'fixing' solution and left for a few minutes before being placed into water, after which the film was air-dried.

2.7IMMUNO - 'DOT BLOTS'

Another method for the detection of specific proteins in a given mixture involved the application of small volumes (usually 1-10 pi) of proteins mixtures onto a

44 nitrocellulose membrane. These protein samples were applied in such small volumes

that they appeared as dots on the membrane (hence, the resulting membrane is known as

a 'dot-blot'). Once the samples had dried, the protein of interest was detected using

immuno-detection (section 2.5) and ECL techniques (section 2.6).

2.8 Bicinchoninic Acid (BCA) MICRO-PROTEINASSA Y

The BCA micro-protein assay (Smith et al, 1985) was used to determine the

concentration of total "unknown" proteins in solution. This technique was performed in

96-well microtitre plates (Sarstedt, SA). The compositions of buffers used in this

procedure are shown in Table 2.4.

Reagents Composition

Reagent A 8% (w/v) Na2C03.H20, 1.6% (w/v) Na2 tartrate, in millie-Q water pH 11.25.

Reagent B 4% (w/v) BCA.Na2 in milli-Q water

Reagent C 4% (w/v) CuS04.5H20 in milli-Q water

Working reagent (WR) 4 Volumes of reagent C, 100 volumes of reagent B, 104 volumes of reagent A.

Table 2.4: Compositions of BCA reagents used for the determination of protein concentrations.

2.8.1 Methods - To a clean 96-well microtitre plate, 100 pi of BSA protein standards were added to the wells in the first three columns (setting up a triplicate series) in order of decreasing concentration (i.e. from 80-0 pg/ml) down the plate. To the next three columns, a series of binary or trinary dilutions of the "unknown" protein sample was performed down the plate (giving a final volume of 100 pi. To each well, 100 pi of WR

(Table 2.4) was added. With a lid in place, the plate was sealed with cellotape and

45 incubated at 60 °C for 30 min; during this time, a purple colour developed in wells containing protein. The plate was then removed and allowed to cool to room temperature, after which the tape and the lid were removed. The plate was subsequently read at 595 nm. A plot of absorbance of the BSA standards was generated and used to interpolate the concentration of "unknown" proteins.

2.9 IMMUNOAFFINITY PURIFICATION OF CLUSTERIN

2.9.1 Preparation of Human Serum - Human serum (Red Cross Blood Bank, Sydney

Australia; stored at -80°C) was thawed in a 37°C water bath before protease inhibitors,

0.2 mM phenylmethylsulfonylfluoride (PMSF; active against chymotrypsin and trypsin, added from a stock solution in DMSO) and 1 mM Ethylenediaminetetraacetic acid

(EDTA; active against metalloproteases), were added. The serum was then suction filtered through a 70 mm glass microfibre filter paper (Whatman).

2.9.2 G7-Sepharose Purification of Clusterin - Before being used, the G7 anti-clusterin monoclonal antibody-coupled Sepharose 4B immunoaffinity column was equilibrated with PBS/azide until the absorbance, as measured by a Bio-Rad Econo UV monitor

(model EM-1) and recorded on a Bio-Rad model 1325 Econo recorder (Figure 2.5), had reached a stable base line. The peristaltic pump was set to a flow rate of 0.5 ml/min; the

UV monitor was set to 0.2 absorbance units, whilst the chart recorder was set at 100 mV with a scroll rate of 0.2 cm/min. The protease-inhibited serum was then passed through the column, followed by PBS/azide until the base line had been restored. Clusterin, at this stage, was immobilised on the column. 0.5% (v/v) Triton X-100 in PBS/azide was then passed through the column to remove any apolipoprotein A-I bound to the immobilised clusterin, followed by more PBS/azide. Once the base line had been

46 reached, clusterin was desorbed and eluted from the column using 2 M guanidine hydrochloride (pH 8.0) in PBS. Lastly, PBS/azide was pumped through the column to wash out excess guanidine hydrochloride and to recondition the column. Clusterin, having been eluted with guanidine hydrochloride (a protein-denaturating agent), was thence dialysed against 5-6 L of PBS/azide at 4°C (overnight).

2.9.3 Protein G Purification of Clusterin - Clusterin is known to bind immunoglobulins

(Wilson and Easterbrook-Smith, 1992). Therefore, when the serum fraction is passed through the G7-sepharose immunoaffinity column, some immunoglobulins may be co-purified with clusterin. Clusterin-Ig fractions (eluted from the G7) were passed through a protein G column (which binds IgG but not clusterin). Since clusterin does not bind the protein G, it was collected as the first peak. PBS/azide was subsequently pumped through the column to restore the absorbance (at 280 nm) to base line. The bound Ig fraction was then eluted (producing the second peak) using 2 M guanidine hydrochloride (pH 8.0), after which, more PBS/azide was used to wash the column and restore A2so to base line. Concentrations of clusterin solutions were determined by BCA micro-protein determination.

47 CD Q « 3 ^3 O s £ 0) o >> 2 C > JO Q i—- ii—iri =o «2 _ |, o o o 6 o o oj L^_ .u2 * § —•73 • . OJ 3 D o ~o o iS C »- Xi i 8 o <+•* e ° c e B o E P a. x. •Ds ^O 03 B G

fll *- •'-' "> g £ "E e « M U o c « fJ •22 XI 60 i o 3 So O c •c o a.• a o s r- o -a O o 3 £ B g — xj C3 B

•"^ ^. •S — M C cS H c -g O ed O u cfl "> -—u- S .3 « 3 y oooo go — "o a V3 *J c- pe'.S e- 2 *•> xs S° § .9 E ° P \s " o u a> < B o _3 .8 C c •a A" H .2 3 o '5 . 03 1/5 a. f4 3 .a x> s & V nl .5 u ^ o •<= s •c

48 Table 2.5: A tabulated list of chemicals and reagents (including their formulae and the name of their supplier) used throughout this thesis project. NAME FORMULAE COMPANY/ SUPPLIER

41D Anti-clusterin Antibody - 78E Anti-clusterin Antibody - oc-Crystallin (total) (From Bovine Eye Lens) - Giftfrom Yok e Berry cc-Lactalbumin (From Bovine Milk) - Sigma

Acetic Acid (Glacial) CH3COOH Ajax Chemicals Acrylamide C3H5NO ICN Biomedicals adenosine 2',5'-Diphosphate (ADP) C10H15N5O10P2 Sigma

Adenosine 5'- Triphosphate (ATP) C,oH14N5OnP3Na2 Sigma Agarose (Analytical Grade) - Promega Alcohol Dehydrogenase (ADH) (From Equine Liver) - Sigma Ammonium Persulfate (NR^S.Og Sigma

Bis-Acrylamide C7H10N2O2 ICN Biomedicals

Bicinchoninic Acid (BCA) C2oH10N204Na2 Sigma Bovine Serum Albumin (BSA) - Sigma Bromophenol Blue Ci9riioBr405S Sigma Casein (from Bovine Milk) - Sigma Catalase (from Bovine Liver) - Sigma

Citric Acid C6H807.H20 Ajax Chemicals Colloidal Coomassie G-250 Staining Solution - NoveLine Pty Ltd

Coomassie Blue (R250) C45H44N307S2Na Sigma

Copper Sulfate CuS02.5H20 Ajax Chemicals Cyanogen Bromide- Activated Sepharose 4B - Sigma Dextran - ICN Biomedicals

di-Sodium Hydrogen Orthophosphate Na2HP04 Ajax Chemicals

Dimethyl Sulfoxide (DMSO) (CD3)2S0 BDH Chemicals

Dithiothreitol (DTT) C4Hio02S2 Sigma DMEM/HAMSF-12 - TRACE Biosciences

Ethidium Bromide (EtBr) C21H20N3Br Sigma

Ethylenediaminetetraacetic Acid (EDTA) Ci0H,2N2O8Na4 Sigma Foetal Calf Serum (FCS) - TRACE Biosciences G7 Anti-clusterin Antibody - - y-Crystallin (From Bovine Eye Lens) - Sigma Glutathione s-Transferase (GST) - Giftfrom Simo n E.B-Smith

Glycerol C,H803 Sigma ICN Biomedicals Glycine (Gly) C2H5N02

Guanidine Hydrochloride (GuHCl) CH5N3.HC1 ICN Biomedicals HEPES ICN Biomedicals Hydrochloric Acid HC1 Ajax Chemicals Ajax Chemicals Hydrogen Peroxide H202 Lysozyme - Sigma Magnesium Chloride MgCl2.6H20 United States Biochemicals 2-Mercaptoethanol (2-Hydroxyethylene Mercapan; beta- C2H60S Sigma Mercaptoethanol) Ajax Chemicals Methanol CH,OH N-(2-Hydroxyethyl) piperazine-N'-(2-Ethanesulfonic acid) (HEPES) C8HlgN204S Sigma Sigma Nicotinamide Adenine Dinucleotide (reduced form beta-NADH) C2iH27N70,4P2

49 Ovotransferrin (alias; Conalbumin) from Chicken Egg White - Sigma

Phenylmethylsulfonylflouride (PMSF) C7H7F02S Sigma

Phosphoenol Pyruvate (PEP) C3H206P Sigma Potassium Chloride KC1 Ajax Chemicals

Potassium di-Hydrogen Orthophosphate KH2P04 Ajax Chemicals

Propidium Iodide C27H34N4I2 Sigma Protein G - Sigma Pyruvate Kinase/ Lactate Dehydrogenase (from Rabbit Muscle) (PK/LDH) - Sigma Sheep Anti-mouse Ig (Affinity Isolated; FTTC-Conjugated) - Silenus Sheep Anti-mouse Ig (Affinity Isolated; HRP-Conjugated) - Silenus

Sodium Azide NaN3 Sigma Sodium Chloride NaCl Ajax Chemicals

Sodium Dodecyl Sulfate (SDS) C12H250S03 BDH Biochemical

Sodium Carbonate Na2C03.H20 Ajax Chemicals

Sodium Hydrogen Carbonate NaHC03 Ajax Chemicals Sodium Hydroxide NaOH Ajax Chemicals Superose-6 Gel Filtration Medium Pharmacia/ Amersham - Biotech.

Tartrate (di-sodium) C4H406Na2.2H20 Ajax Chemicals

TEMED(N,N,N',N'-Tetramethylethylenediamine) C6H16N2 Sigma

Tris- (Hydroxymethylaminomethane) C4H„N03 ICN Biomedicals Triton X-100 - Ajax Chemicals Chapter 3

Clusterin has Chaperone Activity Similar to the Small Heat-Shock Proteins

51

^ 3.1 INTRODUCTION

Like clusterin, small heat shock proteins (sHsps) are abundant and highly ubiquitous proteins. Under physiological conditions, sHsps account for approximately

0.1% of the total protein content in cells. Small Hsps have been detected in virtually all major types of organisms, from bacteria to plants and animals (Jakob et al, 1993;

Jones et al, 1986; Leroux et al, 1997). Whereas only a few members of the sHsp family are present in mammals, birds and yeast, numerous sHsps have been reported in the cytosol and organelles of cells from Drosophila melanogaster and plants

Ehrnsperger et al, 1997; Klemenz et al, 1991). The major eye lens protein, cc- crystallin, which exists in two highly homologous isoforms, ocA- and aB-crystallin, have also been included in the sHsp family due to their structural and functional similarities. However, unlike other Hsp families, members of the sHsp family share very little sequence homology with one another. Despite this, sHsps are grouped together on the basis of (i) a highly conserved C-terminal region, termed the "a- crystallin domain" (ii) their low monomeric molecular mass, (iii) increased expression as a result of heat shock, and (iv) their ability to act as ATP-independent molecular chaperones (Gething, 1997; Lee etal, 1995; Welch, 1985).

Structurally, CD spectroscopy has revealed that sHsps are predominantly (3- structured whilst electron microscopy studies have shown that they exist as globular complexes with a diameter of about 12-18 nm. Clusterin, on the other hand, has been predicted to be predominantly oc-helical (41%), with the remainder of the sequence consisting of p-sheets (19%) and turns (32%). At present, no detailed electron microscopy images are available for clusterin. Whereas one subunit of clusterin has a

52 molecular mass of approximately 40 kDa, the subunit molecular mass of sHsps ranges from 12 to 42 kDa (Jakob et al, 1993). In the cell, sHsps do not exist as monomers, but rather aggregate to form large ring-shaped complexes called heat shock granules that share structural similarities with prosomes and proteosomes (Klemenz et al,

1991). These heat shock granules vary significantly in size and number of subunits between proteins. For instance, the structure of Mycobacterium tuberculosis Hspl6.3 is comprised of a 150-kDa trimer of trimers (i.e. a total of 9 subunits) and is one of the smallest characterised sHsp quaternary structure, whereas the structure of

Methanococcus jannaschii Hspl6.5 has been reported to be comprised of twenty-four monomers that interact with each other to form a hollow spherical complex with an outer diameter of 120A and an inner diameter of 65A (Kim et al, 1998). Furthermore,

Pisum sativum Hspl8.1 and bovine oc-crystallin have 12 and 32 subunits, respectively.

In general, plant sHsp complexes are typically 200-300 kDa in mass, whilst yeast and mammalian sHsps exist as oligomeric complexes of 32-40 subunits with an average total molecular mass of 300-800 kDa (Jakob et al, 1993).

Small Hsps share no sequence homology with clusterin. Human clusterin and bovine ocA-crystallin share only three small regions of sequence similarity. These regions correspond to human clusterin residues 203-210, 227-233, and 243-249.

Residues 243-249 also correspond to a region of the "oc-crystallin domain" (residues

70-76 of ocA-crystallin) (Leroux et al, 1997). Whether this has some structural and/or functional significance remains unknown. However, based solely on the fact that the primary function of the oc-crystallin domain of sHsps is to mediate interactions between sHsps and hydrophobic regions of partially unfolded substrates, it is possible that the corresponding region in clusterin may also be responsible for its ligand binding ability.

This remains to be established.

53 The expression of sHsps, like clusterin, is elevated primarily in response to heat shock. The level of sHsps produced as a result of heat shock can increase by as much as 20-fold, to amount to approximately 1% of the total intracellular protein content

(Waters et al, 1996). In plants and nematodes, this increase may even reach as high as

200-fold (;Jones et al, 1986; Waters et al, 1996). (Courgeon et al,

1988), deprivation, exposure to chemical agents (Stringham and Candido,

1994), infection by pathogens and cellular injuries (Mehlen et al, 1995) may also lead to an increased production of sHsps. As molecular chaperones, sHsps function in an

ATP-independent manner to protect organisms from stress and confer thermotolerance to cells. Recently, clusterin has been shown to act in vitro as a chaperone inhibiting the stress-induced precipitation of a wide range of substrate proteins in the absence of ATP

(Humphreys et al, 1999). In vivo, sHsps have been shown to bind cytoskeletal components - actin and the intermediate filaments, desmin, vimentin, and glial fibrillary acidic proteins. The survival of cells subjected to heat shock and additional oxidative stresses is enhanced due to the sHsps conferring increased stability to the actin filaments (Jakob et al, 1993).

The ability of sHsps to prevent the aggregation of numerous non-related stressed substrates in vitro is well documented (Horwitz, 1992; Jakob et al, 1993;

Leroux et al, 1997; Chang et al, 1996; Kim et al, 1998); oc-crystallin, human and murine Hsp25 were amongst the first sHsps to be identified with chaperone activities (Horwitz, 1992; Jakob et al, 1993). The types of substrates favoured by sHsps are usually identified as being in an intermediately folded (molten globule) state and are highly prone to irreversible aggregation (Treweek et al, 2000; Carver et al,

1995). At present, the underlying mechanism of chaperone actions of sHsps (and clusterin) has not been solved although it is thought that stabilisation of target proteins

54 by sHsps most likely occurs through hydrophobic interactions. This belief is based mostly on data showing increased binding of oc-crystallin to hydrophobic dyes, 8- anilino-1-naphtalene sulfonate (ANS) and bis-ANS, upon elevated temperature; a stress condition which causes oc-crystallin to undergo structural changes, resulting in increased exposed hydrophobicity and increased chaperone action Carver et al, 1995).

In addition, the oc-cystallin domain, which is conserved amongst the sHsps, is believed to be directly involved in mediating the chaperone function of sHsps; this is supported by studies in which mutations, introduced into the oc-crystallin domain of ocB- crystallin, caused partial or complete loss of the ability of ocB-crystallin to inhibit the aggregation of stressed target proteins and to confer thermotolerance in E. coli

(Muchowski et al, 1999; Plater et al, 1999).

An important feature of some sHsps is their ability to stabilise unfolded proteins in such a way that they can be subsequently refolded. This is a crucial mechanism to ensure that non-native proteins, stabilised by molecular chaperones, can be refolded back into their functionally active state and, hence, salvaged for further use.

While some sHsps are capable of partially refolding their own substrates, as demonstrated by Hsp25 and Hsp27 acting on chemically denatured citrate synthase and oc-glucosidase (Jakob et al, 1993), peptides that are stabilised by other sHsps (e.g. Hsp

18.1, a pea sHsp) are refolded by more specialised classes of chaperones, many of which are ATP-dependent (Lee et al, 1997). For example, the E. coli sHsp, IbpB, has been shown to stabilise heat-denatured malate dehydrogenase (MDH) and urea- denatured lactate dehydrogenase (LDH) in folding-competent states for subsequent refolding by DnaK, the major Hsp70 of E. coli, in a process that is accelerated by

Hsp60. It is not known whether clusterin can refold its substrates or indeed, whether it

55 can stabilise unfolded proteins in a state competent for subsequent refolding by other known chaperones with refolding capability.

Two well-documented classes of molecular chaperones involved in in vivo protein folding include Hsp70 and Hsp60. The active form of Hsp70 (Hsp70-ADP) binds to nascent polypeptides as they emerge from the ribosome, thereby preventing inappropriate interactions between exposed hydrophobic areas on 'self and neighbouring protein structures (Garret, R., and Grisham, CM., 1995). The partially folded protein (in the molten-globule state) is then released from Hsp70 in an ATP- dependent step in which ATP is hydrolysed to ADP, before being encapsulated by the

Hsp60/GroEL complex. Here, the polypeptide undergoes a succession of events including protein binding, ATP-dependent release, and progressive folding prior to being released in its biologically active and fully folded state.

Interestingly, the ability of Hsp70 to directly facilitate protein refolding has also been documented. In 1993, Schroder et al. showed that DnaK was capable of refolding denatured luciferase in vitro; whilst, in 1995, Simons et al, 1995, reported that BiP, the major Hsp70 of the endoplasmic reticulum (ER), was able to refold carboxypeptidase

Y (a vacuolar protein) in vivo after translocation through the ER membrane. Similar to

Hsp60-mediated protein folding, ATP-hydrolysis is believed to be crucial for peptide binding, release and refolding by Hsp70.

This chapter examines the role of clusterin as a molecular chaperone during in vitro protein unfolding and refolding. The current study demonstrates that clusterin's extremely broad substrate specificity extends beyond those recently reported by

Humphreys et al, 1999, and includes proteins in undiluted human serum. In addition, results from native gel electrophoresis and size-exclusion chromatography will be

56 presented to show that, like the sHsps, clusterin also forms high molecular weight

(HMW) complexes with its substrates. Furthermore, results will be presented to illustrate the inability of clusterin to protect two proteins from heat-induced loss of activity as well as the apparent failure of clusterin to catalyse the recovery of enzyme activity following the removal of stress. Although clusterin is predicted to contain, within its sequence, a nucleotide-binding motif (Choi et al, 1989), it is not known whether clusterin possesses ATPase activity or whether ATP has any effect on clusterin's chaperone activity. These issues were also investigated. In an attempt to address the fate of clusterin-bound substrates, this chapter concludes by presenting data to show that, like some sHsps, clusterin stabilises heat-stressed enzymes in a manner which allows for subsequent refolding by Hsc70 (the constitutively expressed form of

Hsp70), resulting in the partial recovery of enzyme activity.

57 3.2 METHODS

3.2.1 Inhibition of Protein Precipitation - The ability of human serum-derived clusterin (/JSCIUS) to protect substrate proteins from stress-induced precipitation was examined by measuring the turbidity associated with protein precipitation as absorbance at 360 nm. A variety of assays were performed, (i) Individual solutions of ovotransferrin (1 mg/ml) or total y-crystallin (1 mg/ml) in 50 mM Na2HPC>4 buffer, pH

7.0, or mixtures of one of these proteins (at the same concentration) with /zsClus (25-

400 pg/ml) were heated for 40 min at 60°C. The light scattering associated with protein precipitation for all solutions was measured at 2 min intervals using an automated seven-chambered diode-array spectrophotometer (Hewlett-Packard GMBH,

Germany), (ii) Solutions of ovotransferrin (1 mg/ml) or lysozyme (250 pg/ml) in 50

mM Na2HPC»4 buffer, pH 7.0, or mixtures of one of these proteins (at the same concentration) with /zsClus (100-1000 pg/ml) were incubated in 96 well microtitre plates (Disposable products, Adelaide, Australia) for 6 h with or without the addition of

20 mM DTT. During this time, light scattering of each solution was measured every 8 min with a Spectramax 250 microplate reader (Molecular Devices, Sunnyvale, CA).

(iii) The effects of ATP on the inhibition of stress-induced protein precipitation by clusterin were investigated by monitoring the turbidity associated with target protein precipitation as absorbance at 360 nm. In all cases, the buffer used contained 50 mM

Na2HP04, 50 mM KC1, and 5 mM MgCl2; in the presence or absence of 2 mM ATP.

(A) Solutions of catalase (1 mg/ml) were heated at 60 °C, in the presence or absence of clusterin (100 pg/ml). (B) Ovotransferrin (1 mg/ml) was heated at 70 °C in the presence or absence of 75 pg/ml clusterin. (C) Lysozyme (0.25 mg/ml) was reduced with 20 mM DTT at 42 °C with or without clusterin (400 pg/ml). The extent of catalase and ovotransferrin precipitation was measured using a diode-array

58 spectrophotometer (Hewlett-Packard GMBH, Germany) whereas lysozyme precipitation was conducted in 96 well microtitre plates and measured using a

Spectramax 250 plate reader (Molecular Devices, Sunnyvale, CA). (iv) A solution of human serum was equally divided into two aliquots. One aliquot was left untreated and was subsequently named 'normal human serum' (NHS). The other, named

'clusterin-depleted serum' or CDS, was depleted of clusterin by passing over an anti- clusterin monoclonal antibody immunoaffinity column (as described in chapter 2).

Clusterin depletion was confirmed by dot-blot analysis. In the first series of experiments, both NHS and CDS were diluted 1 in 20 in PBS and supplemented with

100 pg/ml of purified clusterin or control proteins. A time course measuring the amount of serum protein precipitation in these assays was conducted at 60°C using a diode-array spectrophotometer (Hewlett-Packard model 8453 spectrophotometer; integration time, 0.5 s; cycle time, 30 s), over a period of about 16 minutes. The extent of protein precipitation was measured as an increase in absorbance at 360 nm. In the second series of experiments, undiluted NHS or CDS were incubated at 37°C with or without 20 mM DTT, and in some cases, with 100 pg/ml of clusterin or control proteins added. These experiments were performed in 96 well microtitre plates and the turbidity at A360 was measured using a Spectramax 250 plate reader.

3.2.2 ELISA - Native ovotransferrin, at 20 pg/ml in 50 mM Na2HP04 buffer, pH 7.0, was adsorbed onto 96 well microtitre plates for 1 h at 37 °C. In some cases, ovotransferrin was stressed during coating of the wells by either (i) heating the plate at

60 °C for 1 h or (ii) adding 20 mM DTT to the coating buffer and incubating for 5 h at

37 °C. Native lysozyme, at 20 pg/ml in PBS, was adsorbed onto 96 well microtitre plates for 5 h at 37 °C in the presence or absence of 20 mM DTT. In all cases, the

59 plates were subsequently blocked with 1% (w/v) heat-denatured casein (HDC) for 1 h at 37 °C prior to incubation with 5 mM R(+)-[6,7-dichloro-2-cyclopentyl-2,3-dihydro-

2-methyl-l-oxo-lH-inden-5-yl)oxy] acetic acid (IAA) for 1 h at 37 °C, to exclude the possibility of subsequent disulfide bond formation between clusterin and coated proteins. Purified clusterin, initially at 10 pg/ml, was then serially diluted in binary steps down the plates, using 1% HDC as diluent, and incubated at 37 °C for 1 h. To minimise non-specific binding of clusterin to the ELISA plates, three washes were performed with 0.1% (v/v) Triton X-100 in PBS. A cocktail of G7, 78E and 41D anti- clusterin monoclonal antibodies was used to detect bound clusterin; DNP-9 was used as an isotype control. All antibodies were used in the form of unpurified hybridoma culture supernatant. Bound primary antibodies were detected with sheep-anti-mouse-

Ig-HRP (Silenus, Sydney, Australia) using o-phenylenediamine dihydrochloride (OPD;

2.5 mg/ml in 0.05 M citric acid, 0.1 M Na2HP04, pH 5.0, containing 0.03% (v/v)

H202) as substrate.

3.2.3 Sephacryl S300 Size Exclusion Chromatography - Individual solutions of clusterin (1 mg/ml) or ovotransferrin (5 mg/ml) and mixtures of clusterin (1 mg/ml) with ovotransferrin (at 5 mg/ml) were left untreated or treated with heat (60°C for 1 h;

all solutions in 50 mM Na2HP04, pH 7.0) or DTT (20 mM DTT at 37°C for 5 h; all solutions in 50 mM Na2HP04, 0.1 M NaCl, pH 7.0). Similarly, solutions of clusterin

(1 mg/ml) or lysozyme (5 mg/ml) and mixtures of clusterin and lysozyme (at the same

final concentrations; in 50 mM Na2HP04, pH 7.0) were left unstressed or stressed by heating at 60 °C for 1 h in the presence of 20 mM DTT. All samples were then centrifuged (1 min at 10,000 rpm in a benchtop microfuge) to remove precipitated proteins. 50 pi of each solution was loaded onto a 25 x 1 cm Sephacryl S300 column

60 (Pharmacia Biotech, Melbourne, Australia), which was previously equilibrated with

PBS/Az. Separations were performed at a flow rate of 0.25 ml/min using a low- pressure liquid chromatography system equipped with a 280 nm flow cell

(Econosystem; Bio-Rad, Sydney, Australia).

3.2.4 SDS-PAGE Analysis of Clusterin-Substrate Complexes - The HMW complexes formed between (i) clusterin and ovotransferrin undergoing heat- or DTT-mediated stress, or (ii) lysozyme undergoing DTT-mediated stress, eluted in exclusion volume peaks during size exclusion chromatography were analysed by SDS polyacrylamide gel electrophoresis. In each case, 10 pg of purified clusterin, ovotransferrin, or lysozyme, or mixtures of clusterin with one of the other proteins, and proteins collected in the exclusion volume peaks from size exclusion chromatography were subjected to either native agarose gel electrophoresis or SDS-PAGE (see sections 2.2 and 2.3). Proteins that were separated by agarose gel electrophoresis were subsequently transferred onto a nitrocellulose membrane by western transfer procedure (section 2.4), stained and immuno-labelled. Proteins separated by SDS-PAGE were electrophoretically transferred onto PVDF membrane (section 2.4) and immuno-labelled. Ovotransferrin was biotinylated using standard procedures. Immunolabelling was performed using (i) anti-clusterin Mabs followed by sheep-anti-mouse Ig-HRP conjugate (Silenus,

Melbourne, Australia) or (ii) streptavidin-HRP (Amersham, Sydney, Australia), followed by ECL (section 2.5).

61 3.2.5 Assays of Heat-Induced Loss of Enzyme Activity - Loss of ADH enzyme activity was assayed by removing aliquots from samples of enzyme being heated at 55 °C for

10 min. Prior to assay, aliquots of ADH were held undiluted on ice. Precipitated protein was removed by centrifugation (30 s at 10,000 g in a benchtop microfuge) and aliquots of the supernatant added to a reaction buffer (2.5 mM NAD+, 100 mM ethanol

in 50 mM trisCl, 100 mM NaCl, 5 mM MgCl2, pH 7.4). The formation of NADH was measured by monitoring the increase in absorbance of the solutions at 340 nm for 30 s.

In control experiments (not shown), it was established that plots of absorbance vs time were linear for at least 60 s in these assays and that there was a linear relationship between the rate of change of absorbance and the amount of ADH added. Loss of catalase enzyme activity was assayed by removal of aliquots from samples undergoing

heat-induced precipitation at 55 °C and immediately adding them to 0.12% (v/v) H202 in 50 mM phosphate, pH 7.0. Catalase activity was measured as a decrease in absorbance at 210 nm. In some experiments, the enzymes were heated in the presence of 100 pg/ml clusterin and/or 2 mM ATP.

3.2.6 ATPase Assays - Production of ADP from ATP was measured using an enzyme- coupled assay in which ADP production is linked to oxidation of NADH. The reaction

mixture contained 2 mM Hepes, pH 8.0, 10 mM MgCl2, 100 mM KC1, 10 uM EDTA,

170 pM ATP, 840 pM phosphoenol pyruvate, 300 pM NADH, 33.4 units/ml of lactate dehydrogenase and 21.1 units/ml of pyruvate kinase (Lilley and Portis, 1997). The reaction mixture (0.7 ml) was held in a quartz cuvette, maintained at 37 °C, and NADH oxidation monitored as a decrease in absorbance at 340 nm as a function of time after addition of clusterin. The validity of this assay was confirmed by showing that (i) addition of exogenous ADP (21 nmol) and (ii) generation of ADP from ATP through

62 phosphorylation of 3-phosphoglycerate (1.1 pmol added) catalysed by 3- phosphoglycerate kinase (0.76 units added) both led to NADH oxidation. In order to test the possibility that clusterin might exhibit ATPase activity only when bound to stressed proteins, a 180 pi aliquot of a mixture of clusterin (75 pg/ml) and ovotransferrin (1 mg/ml) that had been heated at 60 °C for 5 min was added to the

ATPase reaction mixture and the assay performed as described.

3.2.7 Thermal Denaturation and Renaturation Experiments - To determine whether clusterin is able to facilitate the renaturation of its substrates, thermal denaturation and renaturation of two enzymes, catalase and ADH, was conducted, (i) Solutions of catalase (0.2 mg/ml) in 100 mM Na2HP04, 50 mM KC1, 5 mM MgCl2, pH 7.0, with or without clusterin (100 pg/ml) or lysozyme, a-lactalbumin, or myoglobin (control proteins, each at 100 pg/ml), were heated at 55 °C for 30 min. The solutions were then cooled on ice and centrifuged for 30 s at 10,000 g to remove precipitated protein. A

total of 20 pi of supernatant was added to 180 pi of refolding buffer (5 mM MgCl2, 50 mM KC1, 1 mg/ml BSA, 50 mM Tris, pH 7.5), with or without the addition of 35 pg/ml Hsc70 (the constitutively expressed form of Hsp70) and/or 2 mM ATP, and incubated at room temperature. At various times, aliquots were taken and assayed for catalase activity as above, (ii) Solutions of ADH (1 mg/ml) in 50 mM Na2HP04, pH

7.5 with or without clusterin (100 pg/ml) or lysozyme (control protein; also at 100 pg/ml) were heated at 55 °C for 10 min. The solutions were subsequently cooled on ice and then cleared of precipitated protein by centrifugation at 10,000 g for 30 s. 20 pi of supernatant was added to 180 pi of refolding buffer and incubated at room temperature. At various times, aliquots were taken and assayed for ADH activity. This was achieved by placing the aliquots into a reaction mixture containing 50 mM Tris,

63 + pH 7.5, 100 mM NaCl, 2.5 mM NAD , 100 mM ethanol, 5 mM MgCl2 and 50 mM

KC1; in the presence or absence of 2 mM ATP. The rate of NAD+ reduction was monitored as an increase in absorbance at 340 nm as a function of time.

64 3.3 RESULTS

3.3.1 Clusterin Maintains the Solubility of Stressed Ovotransferrin, y-Crystallin and

Lysozyme.

One of the hallmarks of molecular chaperones is their ability to suppress the aggregation of thermally or chemically denatured substrate polypeptides. A key question to address when considering the role of clusterin as a molecular chaperone is whether it can protect a wide range of unrelated proteins against stress-induced precipitation. A previous report has shown that clusterin inhibits heat-induced precipitation of GST and catalase and DTT-induced precipitation of BSA and a- lactalbumin (Humphreys et al., 1999). In this study, the ability of clusterin to inhibit stress-induced precipitation of three other previously untested proteins, ovotransferrin, y-crystallin and lysozyme, was investigated. These proteins were chosen as model proteins because they were readily available and bear no structural similarities to each other. In the absence of clusterin, the heating of ovotransferrin and y-crystallin at 60

°C produced extensive precipitation within 10 and 40 min, respectively, as measured by light scattering at A360 (Figure 3.3.1, panels A & C). Upon incubation at 42 °C in the presence of 20 mM DTT, extensive precipitation of ovotransferrin and lysozyme occurred within 2 - 3 h (Figure 3.3.1, panels B & D). For DTT-treated ovotransferrin and lysozyme, a small decrease in light scattering was measured commencing at approximately 160 and 180 min, respectively, corresponding to the appearance of large visibly discrete protein aggregates in solution. When exposed to the same stress conditions as test proteins, clusterin precipitation was not observed (data not shown), indicating that clusterin remained soluble under these conditions. When co-incubated with any of the proteins subjected to heat or DTT-mediated reduction, clusterin potently inhibited protein precipitation (Figure 3.3.1).

65 Since clusterin (and the small heat shock proteins) exist in solution as aggregates of varying numbers of subunits, a convention that has been adopted when dealing with the interactions between chaperones and their substrates is to define stoichiometry in relation to the individual subunits of the chaperone and the substrate protein with which it interacts. To determine the stoichiometry of clusterin's chaperone action, it was assumed that the molecular mass for intact clusterin was 80 kDa and the subunit mass, 40 kDa. Of the substrates tested, clusterin was most efficient as a chaperone for heat-

66 stressed ovotransferrin; the subunit molar ratio (SMR) of clusterin:ovotransferrin for complete inhibition of ovotransferrin precipitation was 1.0:10.6 (Figure 3.3.1 A; when ovotransferrin was taken to have a subunit molecular mass of 75.8 kDa). The corresponding SMRs for the complete prevention of precipitation of reduced ovotransferrin (Figure 3.3.IB), y-crystallin (Figure 3.3.1C), and reduced lysozyme

(Figure 3.3.ID), were 1.0:0.96, 1.0:5.0 and 1.0:0.84, respectively. In contrast, control proteins such as ovalbumin had negligible effects on the stress-induced precipitation of the substrate proteins tested (data not shown). These results clearly illustrate the ability of purified clusterin to suppress the in vitro stress-induced aggregation and precipitation of ovotransferrin, lysozyme and y-crystallin in a concentration-dependent manner.

Similar results have also been obtained with heat-stressed ADH (see Poon et al. 2000;

Appendix 1).

3.3.2 Clusterin Protects Serum Proteins from Stress-Induced Precipitation.

The ability of clusterin to protect purified proteins from stress-induced precipitation could also be extended to include proteins found in human serum. When undiluted human serum was heated at 60 °C, an extremely high level of protein precipitation occurred. As a result, it was impossible to measure this event in real time by spectrophotometry. To overcome this problem, both normal and clusterin-depleted human serum was diluted 1 in 20 in PBS before use. When heated to 60 °C, proteins in clusterin-depleted human serum (CDS; diluted 1 in 20 in PBS) precipitated; this was detected and measured as light scattering at A360 (Figure 3.3.2, panel A). Relative to

CDS, the extent of protein precipitation was reduced in similarly treated normal human serum (NHS; diluted 1 in 20 in PBS). This indicates that endogenous levels of clusterin (present in undiluted NHS at 35-105 pg/ml) are sufficient to significantly inhibit protein precipitation induced by heating at 60 °C. The addition of 100 pg/ml of

67 purified clusterin to diluted CDS reduced the extent of heat-induced protein precipitation by about 75% (Figure 3.3.2, panel A). Similarly, the addition of 100 pg/ml of purified clusterin to diluted NHS also resulted in a 75% reduction in stress- induced protein precipitation (Figure 3.3.2, panel A). These results indicate that, relative to the effects of endogenous clusterin and under the conditions tested, supra- physiological levels of clusterin caused greater inhibition of heat-induced protein precipitation. The addition of 100 pg/ml of control proteins (e.g., BSA, ovalbumin) had no significant effects on the extent of heat-induced protein precipitation (data not shown).

68 The extent of protein precipitation induced in undiluted human serum by DTT is much less than that induced by 60 °C heat. Thus, it is possible to follow this event in real time by spectrophotometry. The addition of 20 mM DTT to undiluted CDS resulted in extensive protein precipitation within 40 min after the commencement of the time- course (Figure 3.3.2, panel B). Relative to CDS, the extent of protein precipitation was reduced in similarly treated NHS. The addition of 100 pg/ml of purified clusterin to

CDS reduced the extent of protein precipitation to slightly less than that measured for

NHS. A further decrease in protein precipitation was observed in samples of DTT- treated NHS supplemented with an additional 100 pg/ml purified clusterin (Figure

3.3.2, panel B).

3.3.3 Clusterin Binds Preferentially to Stressed Forms of Ovotransferrin and

Lysozyme.

Clusterin was previously shown to bind preferentially to the stress-denatured forms of numerous proteins including GST, BSA, and a-lactalbumin (Humphreys et al., 1999).

Here, ELISA was used to demonstrate the preferential binding of clusterin (provided either as a purified protein or in unfractionated human serum) to the stressed forms of ovotransferrin and lysozyme (Figure 3.3.3). In all cases, no significant binding of clusterin to unstressed ovotransferrin or lysozyme was observed (Figure 3.3.3 A-C). In contrast, purified clusterin showed significantly increased binding to ovotransferrin, which was stressed with either heat (60 °C; Figure 3.3.3 A) or DTT-mediated reduction

(Figure 3.3.3 B) and to DTT-treated lysozyme (Figure 3.3.3 C). Similar results were obtained when unpurified clusterin, present in whole human serum, was tested (Figures

69 3.3.3 D & E). These results clearly indicate the preferred binding of clusterin to

stressed versus non-stressed proteins.

0 2.5 5.0 7.5 10.012.5

IT) 3,

10000 200 300 Serum Dilution Factor

Figure 3.3.3: Results of ELISA measuring the binding of purified clusterin (A-C) or clusterin in crude human serum (D and E) to native ( • ) or stressed (• ) adsorbed proteins; (A) Heat-stressed ovotransferrin (ovo), (B & D) DTT-reduced ovotransferrin (ovo + DTT), and (C & E) DTT-reduced lysozyme (lys). DNP-9 ( A ) was used as an isotype control antibody for the detection of clusterin binding to stressed proteins. Other details are described in Methods (section 3.2.2). Each data point represents the mean of three replicate measurements and the error bars shown are standard errors (SE) of the mean in each case. In some cases, the SE are too small to be visible. The results shown are representative of three independent experiments.

70 3.3.4 Clusterin Interacts with Stressed Ovotransferrin, Lysozyme and a-Lactalbumin to

Form HMW Complexes.

The ability of clusterin to inhibit the stress-induced precipitation of numerous test proteins, such as those reported in this thesis and in (Humphreys et al., 1999), strongly suggests that it forms stable complexes with these substrates in response to stress. The formation of HMW complexes between clusterin and each of the four stress-affected substrates (GST, catalase, a-lactalbumin, and BSA) was detected by size exclusion chromatography (Humphreys et al., 1999). Here, the complex formation between clusterin and heat- or DTT-treated ovotransferrin, DTT-treated lysozyme, and DTT- treated a-lactalbumin was investigated by several techniques including size exclusion chromatography, native gel electrophoresis and SDS-PAGE.

Size exclusion chromatography was conducted to investigate whether clusterin forms HMW complexes with stressed ovotransferrin and lysozyme. Clusterin, at physiological pH, exists in aqueous solution as an equilibrium mixture of monomers, dimers and higher aggregation states (Easterbrook-Smith, S.B., unpublished results).

There was no significant difference between the elution profiles of untreated versus heat- or DTT-treated clusterin (Figure 3.3.4 A & B), indicating that there was no change in the aggregation state of clusterin in solution in response to heat or reduction by DTT. These results suggest that the quaternary structure of clusterin is stable under these conditions.

Size exclusion chromatographic analyses of protein mixtures subjected to heat- or DTT- mediated stress of ovotransferrin, or DTT-reduction of lysozyme, in the presence of clusterin, revealed the disappearance of the lower order clusterin oligomers and monomers and the concomitant appearance of the HMW species (Figure 3.3.4 A-C). The

71 HMW species were not detected in analysis of clusterin alone or any of the individual proteins, whether these proteins were stressed or not. The HMW species was also absent in unstressed mixtures containing clusterin and ovotransferrin or lysozyme.

When the exclusion volume peak (containing the HMW species), obtained by size exclusion chromatography, of mixtures of clusterin and DTT-treated lysozyme was analysed by SDS-PAGE, both clusterin and lysozyme were detected (Figure 3.3.5). This confirmed that the HMW species was comprised of a complex containing both clusterin and stressed lysozyme (Figure 3.3.5 A). Similarly, when the exclusion volume peaks of mixtures of clusterin and heat- or DTT-treated ovotransferrin were analysed by SDS-

PAGE, both clusterin and stressed target protein were detected, thus verifying that under stress conditions, clusterin binds and forms HMW complexes with stressed proteins.

3.3.5 ATP Does Not Affect the Ability of Clusterin to Suppress Stress-Induced Protein

Aggregation.

The effects of ATP on the ability of clusterin to inhibit the stress-induced precipitation of ovotransferrin, lysozyme, and catalase were investigated and the results presented in Figure 3.3.6. In each case, the addition of ATP to each of the substrates undergoing stress resulted in a small reduction in protein precipitation, which occurred regardless of whether clusterin was present or not. Thus, ATP had no effect on the ability of clusterin to inhibit the stress-induced precipitation of the three proteins tested.

72 u c/3 u- c 0u SE3 V3 •a u £ 3 C r- t- _3 uo CJ Q U CJ VC X w c/i u Q < + + + oj tn 1/5 — O P 5 •t—i >> _>• 5 H C/3 0U U -1 _>, •a H S3 c r- oE Q CJ 03 r<~, (N X ON CJ c 5 X S3 s CJ CJ •a S OJ o c _3 T3 ^H Xi OJ CN 3 (N OS "a CJ Xj C (U E U c t-. c OJ '5 yi 3 OJ o c S3 c 2 c c a S3 _0J OJ X C 'C c •a • p. ••• o ,<4-cu ui« cOJ 2 OJ 'S OJ c C o u ON t/5 S3 5 X o a. J3 u a 73 O 1*- C/3 OJ E aj c S3 o o> >. _0J -a N c CN CJ •a i_ Oin o cS3 c rt IE 3 _3 o ca­ OJ m c OJ p6s0 E > S3 X O \- _3 C •a E o CJ a 'C OJ 0 60 cSJ ^3 eo c C t/3 OJ _u3 X CJ c U 4— C/3 'S E a c C c 60 _3 u U c S3 a "c a 'Tri O o o CJ o-r-j _3 'E OJ ^D 73 > c/CJ] '3 C SC3 c/3 0 CU c o 5 u U u U X (- G o u 0 o o 3 oCJ vD Ci u on X mr- o oO c C/3 '35 1) X c E ON o tn 3 u "E 'C O OJ _3 y] o aj c o OJ GO 73 E >OJ- c fc c25 X CN 7^. i/i o 0 H u C «C/S] S3 u o C 1)r E 3^ S3 S3 E S3 X E u o m Q. N X CJ c O O 0- c on on > S3 S3 •-, _3 OJ OJ "a. O en N "0 _3 s 0 7j a o S3 A LANES 1/3 'a? Q 80 • ^- 50 • CO CO 40 • > rr 25 • < 3 20 • O y 15 • •»•» o 2 10 •

B (i) SDS-PAGE (ii) Immunoblots probed with:

Streptavidin-HRP Anti-clusterin (detected antibodies ovotransferrin) 1 2 3 4 12 3 4 12 3 4 CO Q M * 80^ **~ CO 70 • —'

rr < 40 • _l O 30 • LU _l

Figure 3.3.5: SDS-PAGE gel and immunoblots showing the presence of clusterin and stressed proteins in the HMW complexes obtained in the exclusion peaks fromsize exclusion chromatographic analyses (see Figure 3.4). The positions of molecular markers are indicated by arrows. Approximately 10 |4g of protein was loaded into each lane. (A) Samples electrophoresed on a 12.5% SDS-PAGE gel under reducing conditions: clusterin (lane 1), lysozyme (lane 2), exclusion peak fraction fromthe size exclusion chromatographic separation of a mixture of clusterin and stressed lysozyme (lane 3). (B) (i) Samples electrophoresed on a 10% SDS-PAGE gel under reducing conditions: clusterin (lane 1), ovotransferrin (lane 2), HMW fraction of mixture of clusterin and heat-treated ovotransferrin (lane 3), HMW fraction of mixture of clusterin and DTT-treated ovotransferrin (lane 4). (ii) Immunoblots showing the corresponding fractions as in B (i) probed with either anti-clusterin antibodies or with streptavidin-HRP to detect biotinylated ovotransferrin.

74 (10 0 turbidit y ( • ) , th e clusteri n ovotransferri n an d (0.2 5 mg/ml ) an d

—Fr-§3— ° ) , o f an d # _ stresse d alon e

CO ( lysozym e mixture s wer e mg/ml ) o r o f - CM ( 1 clusteri n an d mg/ml ) catalas e mixture s o f AT P u^^ 1 o Targe t i i i i ^» m M CD CM CO ^ (ovo)( l

0 2

^ ^ 0 C> mixture s bot h CD o r o f precipitation . • 9. fC°O (0.2 5 mg/ml ) o r Ovotransferri n mg/ml ) o (lys ) protei n

- ^t (A )

CM th e presenc i n independen t experiments . (min ) P ^ V O 3.2.1 . Lysozym e o r - CM i n thre e ) , Tim e # Catalas e (cat ) ( 1 o f PQ ^^^ 1 o stressed-induce d c 6 0 °C . (B ) a s describe d D inhibi t . D 1 a t 0.4 - clusteri n ( 36 0 t o 4 2 °C . (C ) A 1.2 - 0.8 - representativ e

CD a t 1 1 a s ar e clusteri n o f

- CO heat-stresse d show n measure d 2 0 m M DT T i n th e presenc o f wer e abilit y wa s ) , result s th e H o n fig/ml ) Th e ( reduce d wit h °C . (7 5 AT P precipitatio n 6 0

i i i i. wer e m M AT P CD CM CO Tf

(uiu 09£) 33UBqjosqv 3.3.6 : presenc e th e mg/ml ) an d |ig/ml ) ( 1 Figur e i n associate d wit h protei n clusteri n

75 3.3.6 Clusterin Does Not Protect Enzymes from Stress-Induced Loss of Function.

Purified clusterin was previously shown to effectively suppress the heat- induced precipitation of two enzymes, ADH and catalase (Humphreys et al, 1999). In this study, the ability of purified clusterin to protect these two proteins against heat- induced loss of enzyme activity was investigated. The results of this experiment are presented in Figure 3.3.7. Regardless of the presence of ATP, in the absence of clusterin, partial (about 30 - 40 %) inactivation of ADH and catalase was achieved by heating at 55 °C for 10 and 30 min, respectively. Under these conditions, clusterin had no effect on the extent of heat-induced enzyme inactivation. The results clearly indicated that, irrespective of whether ATP was present or not, clusterin was unable to protect these two enzymes against heat-induced loss of activity.

u S3 et u © <

mC 0 -H Non- Non- Stressed Stressed Non- Non- Stressed Stressed Stressed Stressed ADH ADH + ATP stressed stressed catalase catalase + ADH ADH + ATP + + catalase catalase + + ATP + clus clus ATP clus clus

Figure 3.3.7: Inability of clusterin to prevent heat-induced loss of enzyme activity. (A) ADH (1 mg/ml) or mixtures of ADH (1 mg/ml) and clusterin (100 pg/ml), in the presence or absence of 2 mM ATP, were heated at 55 °C and then assayed for ADH activity as described in 3.2.7. (B) Catalase (200 pg/ml) or mixtures of catalase (200 pg/ml) and clusterin (100 pg/ml), in the presence or absence of 2 mM ATP, were heated at 55 °C and then assayed for catalase activity as described in 3.2.7. For both experiments, the data (shown as the overall change in absorbance measured) represent means of triplicate measurements; error bars represent standard deviations (SD) of the means. In all cases, the results are representative of at least two independent experiments.

76 3.3.7 Clusterin Has No Detectable ATPase Activity.

The presence or absence of ATPase activity was determined using a

spectrophotometric assay (at A34o) involving NADH oxidation via pyruvate kinase and

lactate dehydrogenase. Purified clusterin (2.44 pM), in the presence of 10 mM MgCl2

and 100 mM KC1, appeared to lack any ATPase activity as evident by the net zero

change in absorbance (measured at A340) over the observed time-course (Figure 3.3.8).

Clusterin also failed to exhibit any ATPase activity when associated with heat-stressed

ovotransferrin (Figure 3.3.8). To verify that clusterin's apparent lack of ATPase

activity was not due to an inefficient assay coupling system, ADP was either added

exogenously or generated from phosphoglycerate kinase-catalysed ATP-dependent

phosphorylation of 3-phosphoglycerate. In these cases, the initial rate of NADH

oxidation became one to two orders of magnitude higher than the control rate,

confirming the validity of the assay

o

s I © < —I 1 1 1 1 1— 0 2 4 6 8 10 12 14 • Time (min) Figure 3.3.8: Spectrophotometric time course measurements showing that clusterin lacks measurable ATPase activity. The absorbance of ATPase reaction buffer (see section 3.2.6) was measured as a function of time after the addition of either 195 pg/ml clusterin ( • ) or a mixture of clusterin and ovotransferrin (previously heated at 60 °C for 5 min) ( A ) to give afinal concentratio n of 15.3 pg/ml clusterin and 0.2 mg/ml ovotransferrin. In both experiments, 21 nmol of ADP was added 11 minutes after the start of the assay (red arrow) to confirm that the assay was functional. In similar experiments, purified ovotransferrin alone was added to the ATPase reaction buffer. No ATPase activity was measured associated with ovotransferrin (data not shown). Shown on the same plot, the absorbance of ATPase reaction buffer (containing 84 pM NADH) was measured as a function of time after the addition of 0.76 units of 3-phosphoglycerate kinase ( # ); at 4 min, 1.1 umol of 3-phosphoglycerate was added (green arrow). The data shown are representative of three independent experiments.

77 3.3.8 Clusterin is Unable to Independently Promote Reactivation of its Substrates but it can Stabilise Stressed Proteins in a State Competent for Refolding by Hsc70.

The ability of clusterin to (i) independently facilitate the recovery of functional activity of two enzymes, ADH and catalase, or (ii) stabilise these substrates in a state competent for refolding by other known chaperones with refolding capability, was investigated. The results of these studies are presented in Figure 3.3.9. Heat- inactivation of ADH or catalase was achieved by heating at 55 °C for 10 and 30 min, respectively, in the presence or absence of purified clusterin. After this treatment, approximately 66% of the original catalase activity remained whereas for ADH, this figure was closer to 82%. Reactivation of these two enzymes was measured as described in section 3.2.7. When acting alone, clusterin was unable to promote the recovery of ADH or catalase activity following the removal of stress (Figure 3.3.9).

An attempt was made to investigate whether clusterin could stabilise stressed proteins in such a way that refolding could be achieved by other chaperones. Unfortunately, at the time these experiments were performed, the absence of any known extracellular mammalian chaperones with refolding capability meant that it was impossible to test for their ability to refold stressed proteins stabilised by clusterin. Therefore, the only alternative was to use an intracellular chaperone with known refolding capability; one such protein was Hsc70 (a constitutively expressed mammalian heat shock protein). In the presence of both Hsc70 and ATP, clusterin-stabilised ADH and catalase exhibited a gradual recovery of the original enzyme activity when measured over a period of 5 hours. For ADH, a recovery of approximately 14% of the original activity was observed (Figure 3.3.9 A); for catalase, the percentage of recovery was approximately

16% (Figure 3.3.9 B). When these enzymes were stressed in the presence of clusterin and then assayed in the presence of Hsc70 but in the absence of ATP, the level of

78 reactivation was less than 4% (Figure 3.3.9). In similar experiments, when lysozyme

was substituted for clusterin, and the heat-stressed enzyme mixture subsequently

incubated with Hsc70 and ATP, the levels of reactivation for ADH and catalase were

approximately 1 and 3%, respectively.

A A ' 16 - A 16 • B

o 12 - 12 • / ADH 8 - 8 • cat Activit y 4 • 4 • Recover y % Enym e 0 i4 "V 1 —r 1 T 1 rr ,—i— 0 1 2 3 4 5 6> o i0 12 3 4 5 6 ^ w Time (h)

KEY + clus -Hsc70 +ATP A -i-clus + Hsc70 + ATP # - clus +Hsc70+ATP O -i-clus + Hsc70 -ATP • + Iyso + Hsc70 + ATP 0

Figure 3.3.9: Post-stress refolding of clusterin -stab 0 is ed ADH and catalase by Hsc70. Solutions of ADH (A) or catalase (B) were heat-inactivated at 55 °C for 10 and 30 min, respectively, in the presence or absence of clusterin (clus; 100 pg/ml) or lysozyme (lyso; 100 pg/ml). Aliquots, cleared of precipitated protein by centrifugation, were then added to refolding buffer, with or without Hsc70 (35 pg/ml) or 2 mM ATP (see key). The recovery of enzyme activity was then measured spectrophotometrically (see 3.2.7). The recovery of enzyme activity is expressed as a percentage of the original enzyme activity (prior to heat stress). The data are means ± standard deviations of triplicate measurements (in some cases the error bars are too small to see) and are representative of three independent experiments. In similar experiments, when a-lactalbumin or myoglobin were substituted for clusterin, results similar to those for lysozyme were obtained (data not shown).

79 Taken together, these results indicate that clusterin does not independently refold heat-stressed ADH and catalase but binds to the stressed proteins to stabilise them in a conformation that can be subsequently refolded by Hsc70. The level of recovery of enzyme activity recovered in these assays is very similar to that reported for recovery of citrate synthase activity following heating with the small heat shock protein Hsp25 and subsequent incubation with Hsp70 (about 14% recovery)

(Ehrnsperger et al, 1997). The effect of clusterin appears to be specific for clusterin because, in similar experiments, a-lactalbumin or myoglobin were unable to support the high level of enzyme reactivation achieved by clusterin (data not shown).

80 3.4 DISCUSSION

The experiments described above extend previous results that had highlighted the ability of clusterin to prevent stress-induced precipitation of a number of unrelated proteins, including glutathione-S-transferase (GST), catalase, a-lactalbumin and bovine serum albumin (BSA). In this project, purified clusterin was shown to inhibit the precipitation of heat-stressed ovotransferrin and y-crystallin, as well as DTT-reduced ovotransferrin and lysozyme (Figure 3.3.1). In addition, for the first time, the ability of clusterin (either purified or in unfractionated human serum) to inhibit stress-induced precipitation of endogenous human serum proteins was demonstrated. Furthermore, by

ELISA, clusterin was shown to bind preferentially to stressed forms of either ovotransferrin or lysozyme. As mentioned in section 1.2.4, the term molecular chaperones was applied by John Ellis to describe a diverse group of functionally related proteins that function in vivo to assist in protein assembly through their stabilising interactions with partially unfolded protein conformers (Ellis, 1987; Ellis, 1992).

Therefore, based on the definition of what a molecular chaperone is, the ability of clusterin to bind preferentially to stressed forms of a wide range of proteins to inhibit their precipitation confirms that clusterin, like the Hsps, is a molecular chaperone.

Clusterin's ability to stabilise numerous stressed proteins (i) independently of ATP and

(ii) in a state competent for post-stress refolding and reactivation by Hsc70 further illustrates functional similarities between clusterin and some sHsps.

One of the general features of molecular chaperones (including the sHsps) is that they interact with the hydrophobic regions of partially folded, unfolded or disassembled polypeptide chains to form solubilised high molecular weight (HMW)

81 complexes. For molecular chaperones such as the mammalian sHsps, these complexes

(typically 300 - 800 kDa in size; Fersht et al, 1994), have been detected at times larger than 1000 kDa (Jenne et al, 1991). Previously, clusterin was shown to form HMW complexes with (i) heat-stressed GST or catalase, and (ii) reductively-stressed a- lactalbumin and BSA. Analysis by size exclusion chromatography of samples in which clusterin was co-incubated with ovotransferrin or lysozyme undergoing stress-induced denaturation revealed the presence of HMW species in the void volume that was eluted from the column (Figure 3.3.4 A,B & D). Subsequent analysis by SDS-PAGE of these

HMW species, confirmed the presence of both clusterin and the stressed protein

(Figure 3.3.5). Since the HMW complexes were absent in samples of clusterin or target proteins alone, whether these were stressed or not, or in mixtures of clusterin with each of the individual substrate proteins in the absence of stress, the results presented here confirm that clusterin interacts preferentially with stressed forms of target proteins only.

To determine the stoichiometry of chaperone action as well as the efficiency of clusterin as a molecular chaperone, potential substrate proteins, e.g. ovotransferrin, y- crystallin and lysozyme, were stressed in the presence of increasing amounts of clusterin and the turbidity associated with substrate precipitation was measured as an increase in absorbance at 360 nm (Figure 3.3.1 A-D). The stoichiometry of chaperone action by clusterin was defined in terms of the subunit molar ratios (SMR) of clusterin to substrate required to completely inhibit precipitation. Table 3.4.1 summarises the physical characteristics, tested stress conditions and SMRs required for the complete inhibition of stress-induced protein precipitation for all target proteins used to-date to assess clusterin's chaperone activity. The format of this table is such that the target

82 proteins are listed in order of decreasing clusterin efficiency. From this table, it can be seen that of all substrates tested, clusterin was most efficient at stabilising DTT-treated insulin; the clusteriminsulin SMR for this was 1.0:50. This was followed closely by the stabilisation of DTT-reduced a-lactalbumin and heat-stressed ovotransferrin with

SMR values for clusterin:substrate of 1.0:11.0 and 1.0:10.6, respectively. The SMR of clusterin:substrate for the other heat- or DTT-treated substrate proteins were much lower, ranging from 1.0:5.0 (clusterin:heat-stressed y-crystallin) to 1.0:0.84

(clusterin:DTT-treated lysozyme) (Table 3.4.1). There does not appear to be any correlation between the SMRs required for the complete inhibition of substrate precipitation and the mass of these substrates, nor does there seem to be any relationship between clusterin's efficiency as a molecular chaperone and the type of stress used. Therefore, how well clusterin functions as a molecular chaperone seems to depend not on the size of the substrate to which it binds or the stress conditions to which its substrates are subjected, but on some unknown characteristic(s) of each of the substrates tested. It is worth noting that clusterin is a more efficient molecular chaperone than the sHsps; existing data suggest that at best, sHsps can only achieve an

SMR of one subunit of sHsp to one partially unfolded substrate protein (Pauling and

Corey, 1951; Anfinsen and Haber, 1961).

83 E & "a vO in XJ * C m ON 00 3 O d o m O c (S3 •— i/S rn CN d d d d d d X3 o on c d d cu •c C _3 O u 73 cj U •a •a ~ p _3 u TJ cj cu 3 •S -a T3 •s c C o a ^3 o CJ CJ CJ 3 3 3 c D. 3 T3 c3 cu cu cu •a T3 T3 c 00 cu CU CJ o s -a cu 3 on <*> on o O 1 ^ CN c e ° o s ^ 2 _. 00 si oo a. re £3 X) on r^ in c 1/S — r~- Q. 3 [23 c 3 i> X) < CU ^

U S •* -H — •* — cd 3 Si 3 CJ Xi H 53 e cs O CN "3 u CJ 1) Q. CO s 00 CO .8 Si SJ —. 00 r- o I Q. CO -a 0CJ5 .5 Q lO ~ 2 58 CN u-i <* CcJ OJ a. r- ir> ^ ^ L4-E1 XI o < 2 3 c .£ on £ cu .S Xc)u 1c cu oCZn iC3S JCZ3 ft TO rrt _032 G-)a cu u — c _3 > wa _cg 3 3 _3 3 cd P T^ Tt T^ " T? rt\ T^ fli, fll .C a O 3 o CU CJ cu 1 <2 2 ST c/3 ci p CJ 53 C/3 CU u < 3 o O O £ Q on c3 > CJ u aes .s oq o 2a "M H <

84 Up to now, many of the experiments investigating clusterin's chaperone action

have been performed in non-physiological buffers (such as 50 mM Na2HP04, pH 7.0).

It was desirable to determine whether the chaperone action of clusterin also operates under physiologically relevant conditions, for example, in serum. As a normal component of human serum, one would expect that if clusterin truly functions in vivo as a molecular chaperone then it should be able to exert a chaperone effect on stressed serum proteins. Results presented in this report demonstrate that clusterin protects proteins in (i) diluted human serum from heat-induced precipitation and (ii) undiluted human serum from DTT-mediated precipitation (Figure 3.3.2). initial experiments investigating the chaperone action of clusterin on heat- induced serum protein precipitation were conducted in undiluted serum (data not shown). However, upon performing this experiment in full sera, a problem was encountered. Serum contains high levels of protein, about 75 mg/mL (Jenne and

Tschopp, 1992). Therefore, when full sera was heated, the rate and extent of protein precipitation was so great that measurements in real time of chaperone effects of clusterin were impossible. To solve this problem, a similar set of experiments was performed using a 1 in 20 dilution of normal human serum (NHS), clusterin-depleted serum (CDS), or either of these supplemented with 100 pg/ml purified clusterin. Heat- and reduction-induced precipitation of serum proteins was (i) enhanced by immunoaffinity depletion of clusterin from serum and (ii) specifically inhibited by the addition of purified clusterin (Figure 3.3.2). No attempt was made to identify the serum proteins that were stabilised by the chaperone action of clusterin. Nevertheless, these findings raise the possibility that clusterin may play a clinically important role in protecting human plasma proteins from stress-induced precipitation. Furthermore, certain diseases such as Alzheimer's and Creutzfeldt-Jacob diseases are associated with

85 abnormally high levels of protein precipitation. In cases such as these, the results presented in this chapter suggest the possibility that the levels of clusterin in biological fluids such as plasma may affect the rate or extent of disease progression.

Clusterin has been predicted to contain, within its amino acid sequence, a putative nucleotide-binding motif [89]. Nucleotide-binding motifs have been reported in many chaperones, such as those belonging to the HsplOO, Hsp70 and Hsp60 heat shock protein families. It is believed that the binding of nucleotides (ATP or ADP) to these chaperones influences the kinetics of substrate binding and/or release (reviewed in Gething, 1997). More importantly, it has been shown that, in most cases, the processes of protein folding or refolding are ATP-dependent and that many of the chaperones with refolding capability are also ATPases. In this study, a number of experiments were conducted to determine (i) whether ATP affects clusterin's chaperone action, (ii) if clusterin has any ATPase activity, and (iii) if clusterin could, alone or in co-operation with other chaperones, refold its substrates.

Results obtained in this chapter have provided evidence to show that, like the sHsps, clusterin does not have the ability to hydrolyse ATP and hence, performs its chaperone function in an ATP-independent manner (Figure 3.3.8). Previous experiments showing the dose-dependent inhibition of stressed-induced precipitation of numerous test proteins (listed in Table 3.1) by clusterin were all performed in the absence of ATP (Figure 3.1; and in Humphreys et al, 1999). Here it was shown that

ATP has no effect on clusterin's in vitro chaperone action (Figure 3.3.6). However, to exclude the possibility that clusterin may possess ATPase activity only when bound to a substrate, clusterin was co-incubated with ovotransferrin undergoing heat-stress prior to being assayed for ATPase activity (see 3.2.6). Once again, results showed that clusterin, even when complexed with a bound substrate, did not exhibit any ATPase

86 activity (Figure 3.3.8). Although the functional domains of clusterin are not known, based on the results obtained here, it appears that clusterin does not contain a functional ATPase domain. In addition, due to the inability of clusterin to hydrolyse

ATP, clusterin was not expected to have the ability to refold partly unfolded protein structures since most chaperones that are able to refold stabilised substrates require

ATP to do so (Preville et al, 1999). This inference was supported by results showing that clusterin was unable to independently facilitate the reactivation of heat-inactivated

ADH and catalase after the removal of stress. However, clusterin was able to stabilise stressed ADH and catalase in a state competent for subsequent refolding by Hsc70 (in the presence of ATP) (Figure 3.3.9).

Taken together, the findings presented in this chapter illustrate many functional similarities between clusterin and the sHsps. Specifically, results show that clusterin

(i) functions in an ATP-independent manner to inhibit the stress-induced precipitation of a variety of unrelated target proteins, including human serum proteins, by forming soluble HMW complexes with them, (ii) does not possess any detectable ATPase activity, (iii) does not protect enzymes from stress-induced loss of function nor independently facilitate their refolding, and (iv) like some sHsps, stabilises target proteins in a state competent for subsequent refolding by an ATP-dependent chaperone.

A subset of the results presented in this chapter have been published in the following journal paper:

Poon, S., Easterbrook-Smith, S.B.,Rybchyn, M.S., Carver, J.A., and Wilson, M.R. (2000) "Clusterin is an ATP-independent chaperone with very broad substrate specificity that stabilizes stressed proteins in a folding-competent state" Biochemistry: 39, 15953-15960. (Appendix)

87 Chapter 4

Clusterin is a pH-Dependent Chaperone Which Specifically Interacts With Disordered Molten Globule States of Proteins

88 4.1 INTRODUCTION

So far, the data presented in Humphreys et al, 1999, and in chapter 3 demonstrate the in vitro sHsp-like chaperone action of clusterin. Specifically, clusterin was shown to prevent the stress-induced precipitation of numerous target proteins, either in the form of purified proteins or in undiluted human serum. Like many sHsps, clusterin does not have the ability to independently refold proteins but stabilises these in a state competent for subsequent refolding by other ATP-dependent chaperones (e.g.

Hsc70; see section 3.3.8). However, similarities between clusterin and the sHsps are not just confined to their functional role as molecular chaperones; both clusterin and sHsps are highly ubiquitous and their expression is up-regulated in response to stress

(e.g. heat shock). In addition, increased expression of both types of chaperone has been detected in many diseases (e.g. Scrapie, Alzheimer's and Creutzfeldt-Jakob diseases) associated with abnormally high levels of misfolded and/or precipitated proteins. Yet, in contrast to the sHsps, information regarding the mechanism of clusterin's chaperone action is lacking. It is unknown what conformational state(s) of substrate proteins clusterin specifically interacts with, or where along the unfolding pathway this interaction occurs. In addition, it is not known what effect certain environmental or pathophysiological factors (e.g. temperature) have on (i) the oligomerisation state of clusterin, (ii) the ability of clusterin to interact with stressed proteins, or (iii) clusterin's chaperone action. This chapter will address these issues.

A number of conditions (generally referred to as "stress"; section 1.2.3) can lead to protein unfolding. As proteins traverse along the unfolding pathway towards the unfolded state, most will undergo a structural transition from the fully folded native state to an intermediate semi-compact molten globule state (section 7.2.2) which,

89 depending on the degree of structure, can be classified as being "ordered" (e.g.

designated Ii) or "disordered" (I2) (Figure 4.1.1). Being partly structured, proteins in the molten globule state expose significant hydrophobicity to solution. This heightened level of hydrophobicity promotes protein - protein interaction, causing the partially unfolded proteins to deviate from the unfolding pathway along the irreversible off-folding pathway, which ultimately leads to their aggregation and precipitation

(Figure 4.1.1).

Information about the dynamics of the molten globule state has been acquired from numerous studies of a model protein called a-lactalbumin (Horwitz et al, 1998;

Griko and Remeta, 1999; Lindner et al, 1997; Lindner et al, 1998). a-Lactalbumin is a small (-14 kDa), monomeric calcium-binding milk protein, which forms a molten globule at low pH or in the absence of calcium ion at neutral pH at temperatures above

25 °C (Griko and Remeta, 1999). Reduction of the four disulfide bonds of a- lactalbumin also results in the formation of a molten globule. The formation of the molten globule is evidenced by a significant reduction in the NMR chemical shift dispersion and a substantial loss of the near-ultraviolet circular dichroism (CD) signal, indicating that the side-chains are substantially disordered. Small Hsps, such as a-

crystallin, have been shown to interact specifically with the disordered (i.e. I2) molten globule state of a number of proteins (including a-lactalbumin) to prevent these proteins from deviating onto the off-folding pathway and to maintain their solubility

(Figure 4.1.1) (Lindner et al, 1997). However, the conformational state(s) of proteins that interact with clusterin and the stage(s) along the unfolding pathway where this interaction occurs have not yet been identified.

90 .s 2 G OH • 2

1- J^ TQ3 0>CN mm\f

!H ~ cu X> SO cu o a « £ m ^ Xi e o cu -a & 0CJ0 o -5 cd -a o >1 > o X3) X) fj -J T3 o OaJ jg •a 9- 13 bfi c a — — CJ cj cj 60 cu o CJ g cu a. oo <2 -5 CJ & S c I-I 00 2 o cj CJ .5 cd -a c £ -a cd CJ M cd u -= cd c -S Q3 .S cu XJ CJ .?C _ ,y u I 8 BO cd £ .s O c « -5 • o < 0c0 CU _cd • cu cj < ^ « 1) a. .2 01 3 t BO XI c 2 u c cd S 3 .2 ™-5, 3 cd E a. c TD cj ^ 3 8

91 Both sHsps and clusterin have been shown to inhibit the slow precipitation of numerous target proteins (such as those listed in Table 3.1). However, sHsps such as a-crystallin are incapable of preventing rapidly aggregating proteins from precipitating

(Lindner et al, 2001). This suggests that kinetic factors may be important in regulating the chaperone action of sHsps. At present, the ability of clusterin to stabilise rapidly precipitating proteins is not known.

For many sHsps, elevated temperature can have a significant effect on their quaternary and tertiary structures. For example, when mammalian a-crystallin is exposed to temperatures above 50 °C, the protein partially unfolds and undergoes a transition to a molten globule-like state, resulting in higher solvent accessibility to hydrophobic surfaces (determined by bis-ANS fluorescence) and an enhancement of chaperone activity (Raman and Roa, 1997; Data and Roa, 1999). Similarly, at physiological temperature, IbpB (an E. coli sHsp) exists in solution as a large multimer of approximately 600 kDa. However, when exposed to elevated temperatures as high as 55°C, IbpB undergoes (i) a partial loss of secondary structure, (ii) dissociation of the multimeric structure into constituent oligomers, and (iii) an increase in protein surface hydrophobicity, resulting in enhanced chaperone action (Shearstone and Baneyx,

1999). Similar effects have also been shown for yeast Hsp26 (Haslbeck et al, 1999).

Like many sHsps, clusterin also exists in solution as a mixture of multimers (see below). However, the effect of elevated temperature on the oligomerisation state of clusterin is not known. In addition, it is not known whether increased temperature has any effect on the ability of clusterin to bind to its substrates or to its chaperone action.

These issues were investigated.

As mentioned in chapter 1 (section 1.3.2), clusterin is capable of aggregating with itself under conditions of neutral pH and low salt concentrations to form

92 dimers and tetramers (Griswold et al, 1986). A recent report has demonstrated that this process is pH-dependent with dissociation of the oligomers occurring at low pH (pH 6.0) (Hochgrebe et al, 2000). Dissociation of clusterin oligomers at mildly acidic pH in vitro resulted in a concomitant increase in the ability of clusterin to bind its native ligands, heparin (Pankhurst et al, 1998), IgG, complement protein C9, apoA-I, and GST (Hochgrebe et al, 2000). Hydrophobic regions exposed as a result of pH-dependent dissociation of clusterin homo- oligomers were thought to be responsible for the increased binding. These results suggest that at low pH, the binding of clusterin to stressed proteins and its chaperone action might also be enhanced. This possibility was investigated.

The effects of temperature and pH on both the interaction of clusterin with target proteins and its chaperone action were examined by ELISA and precipitation assays. In addition, the ability of clusterin to inhibit the rapid precipitation of the target proteins, y-crystallin and lysozyme, was tested. Furthermore, the interaction of clusterin with intermediately folded forms of a-lactalbumin was examined by real-time

'H NMR spectroscopy. The results demonstrate that (i) mildly acidic pH significantly enhances the ability of clusterin to bind to target proteins and enhances its chaperone activity, (ii) temperature does not induce significant changes in the oligomerisation state of clusterin, (iii) clusterin is incapable of inhibiting very rapid protein precipitation, and (iv) clusterin does not significantly alter the rate of a-lactalbumin

reduction but does stabilise the less ordered form (i.e. the I2 state) of the protein.

93 4.2 METHODS

4.2.1 Size Exclusion Chromatography - To determine the effect of elevated temperature on the oligomerization state of clusterin, 100 pi samples of 1 mg/ml clusterin in 50 mM

Na2HP04, pH 7.0, were incubated at 37 or 50 °C for 1 h and immediately applied to a

1.5 x 30 cm (diameter x height) Superose-6 (Pharmacia) column equilibrated with lx

PBS. Chromatography was performed at the indicated temperatures and at a flow rate of 0.5 ml/min. Standards used for calibrating the column included the following: aldolase (158 kDa), catalase (232 kDa), ferritin (440 kDa), (669 kDa) and Blue Dextran (2000 kDa). Kaleidagraph Version 3.51 (Synergy Software,

Reading, PA) was used for quantitative data analysis and plotting.

4.2.2 ELISA - (i) ELISA experiments were used to investigate whether temperature has any effect on the ability of clusterin to bind stressed substrate proteins. GST, at 20

pg/ml in 50 mM Na2HP04, pH 7.0 was adsorbed (whilst being stressed) onto a microtitre plate (Diposable Products, Adelaide, Australia) by incubating at 60 °C for 1 h. BSA and lysozyme (both at concentrations of 20 pg/ml) and ovotransferrin (10 pg/ml) were incubated at 42 °C for 5 hours in the presence or absence of 20 mM DTT.

The plates were then blocked with 1% (w/v) heat-denatured casein (HDC) prepared in

PBS (137 mM NaCl, 2.7 mM KC1, 1.5 mM KH2P04, 8 mM Na2HP04, pH 7.4).

Thereafter, the plate was washed thrice with lx PBS. Plate-bound, DTT-treated proteins were incubated with 5 mM R(+)-[6,7-dichloro-2-cyclopentyl-2,3-dihydro-2- methyl-l-oxo-lH-inden-5-yl)oxy] acetic acid (IAA) for 1 h at 37 °C in order to exclude the possibility of subsequent formation of disulfide bonds between clusterin and the

target proteins. Clusterin, initially at 10 pg/ml in 50 mM Na2HP04, pH 7.0, was then

94 serially diluted in binary steps down each ELISA plate, which was then incubated at

37, 42, 45 or 50 °C for 1 h. A mixture of G7, 4ID and 78E anti-clusterin monoclonal antibodies was used to detect bound clusterin. All antibodies were used in the form of unpurified hybridoma culture supernatant. Bound primary antibodies were detected with sheep anti-mouse Ig-HRP (SaMIgG-HRP; Silenus, Sydney, Australia) using o- phenylenediamine dihydrochloride (OPD; 2.5 mg/ml in 0.05 M citric acid, 0.1 M

Na2HP04, pH 5.0, containing 0.03% (v/v) H202) as substrate.

(ii) To investigate the effects of pH on the binding of clusterin to stressed proteins,

ELISA experiments were conducted as follows. GST and ovotransferrin, at 20 pg/ml in

50 mM Na2HP04, pH 7.0 were adsorbed onto a microtitre plate (Disposable Products,

Adelaide, Australia) by incubating at 60 °C for 1 h. a-lactalbumin and BSA (both at

20 pg/ml) and ovotransferrin (10 pg/ml) in 50 mM Na2HP04, pH 7.0, were incubated for 5 hours in the presence or absence of 20 mM DTT. The plates were then blocked with 1% (w/v) heat-denatured casein (HDC) prepared in PBS (137 mM NaCl, 2.7 mM

KC1, 1.5 mM KH2P04, 8 mM Na2HP04, pH 7.4). The plate was washed thrice with lx

PBS. Plate-bound proteins were incubated with 5 mM IAA for 1 h at 37 °C. Clusterin,

initially at 10 pg/ml in 50 mM Na2HP04, pH 6.0, 6.5, or 7.0, was then serially diluted in binary steps down each ELISA plate, which was then incubated at 37 °C for 1 h.

Bound clusterin was detected with G7, 41D and 78E anti-clusterin monoclonal antibodies, and bound primary antibodies were detected with SaMIgG-HRP and OPD substrate as described above.

4.2.3 Protein Precipitation Assays - (i) The effects of different temperature on the extent of DTT-induced protein precipitation in the presence or absence of clusterin were tested. The turbidity associated with protein precipitation was measured as

95 absorbance at 360 nm. In experiments dealing with the substrate proteins BSA, a- lactalbumin and lysozyme, the extent of protein precipitation was compared at 37 °C and 50 °C. In a separate experiment, insulin precipitation was conducted at 37, 45 and

50 °C. Individual solutions of BSA (750 pg/ml), a-lactalbumin (750 pg/ml), lysozyme

(250 pg/ml) or insulin (2 mg/ml) or mixtures of these proteins with clusterin (at specific concentrations, as indicated in Figure legends), were prepared in 50 mM

Na2HP04, pH 7.0, and reduced with 20 mM DTT at one of the above-mentioned temperatures. Readings were taken every 10 min for a total of 150 or 300 min with a

Spectramax 250 microplate reader (Molecular Devices, Sunnyvale, CA, USA).

(ii) The effects of pH on the inhibition of stress-induced protein precipitation by clusterin was measured as absorbance at 360 nm. In all experiments the extent of precipitation at pH 6.5 was compared to that at pH 7.5. Experiments involving ovotransferrin were also performed at pH 6.0. Individual solutions of ovotransferrin (2 mg/ml) or catalase (300 pg/ml), GST (300 pg/ml) or mixtures of these proteins with clusterin (at specific

concentrations, as indicated in Figure legends), were prepared in 50 mM Na2HP04, pH

6.0, 6.5 or 7.5, where appropriate, and heated at 60 °C. Readings were taken every minute for a total of 30 min using an automated diode-array spectrophotometer (Hewlett-Packard

GMBH, Germany). In a similar experiment, normal human serum (NHS) or clusterin- depleted human serum (CDS; prepared by immunoaffinity chromatography), were diluted

1 in 10 in 50 mM Na2HP04 (pH 6.0-7.5), heated at 60 °C, and measurements of turbidity taken as previously described. Lastly, the extent of protein precipitation at pH 7.5 was compared to that at lower pH values, in the presence and absence of clusterin or the sHsp a-crystallin (purified from bovine eye lenses; a gift from Ms Yoke Berry). Ovotransferrin

(1 mg/ml) or mixtures of ovotransferrin (1 mg/ml) and a-crystallin (650 pg/ml) were

prepared in 50 mM Na2HP04, pH 6.0, 6.5, 7.0, or 7.5, and heated at 60 °C.

96 (iii) The effects of clusterin on slowly precipitating y-crystallin or lysozyme was

determined by heating y-crystallin (1 mg/ml in 50 mM Na2HP04, pH 7.0) at 60 °C in the presence or absence of 50 pg/ml clusterin. Similar experiments were conducted in

which lysozyme (250 pg/ml in 50 mM Na2HP04, pH 7.0) was incubated at 42 °C with

20 mM DTT in the presence or absence of 400 pg/ml clusterin. The extent of protein precipitation in each case was measured as turbidity at 360 nm using a model 8453 diode-array spectrophotometer (Hewlett-Packard GMBH, Germany). To determine what effect clusterin had on rapidly precipitating y-crystallin or lysozyme, the following two assays were performed. (1) 300 pg of y-crystallin was denatured for 20 min in 20 pi of denaturing buffer (50 mM phosphate buffer containing 6 M guanidine hydrochloride (GuHCl)). This solution was added to 10 pi of PBS containing 15 pg of clusterin or one of two control proteins (BSA or ovotransferrin). Rapid y-crystallin precipitation was achieved over 1 min via repeated additions of 10 - 20 pi aliquots of

50 mM Na2HP04, pH 7.0, to give a final volume of 300 pi. (2) 3 mg of lysozyme was denatured in 30 pi of denaturing buffer for 20 min. A 1 pL aliquot (containing 100 pg of lysozyme) was withdrawn and added to 19 pi of PBS containing 160 pg of clusterin,

BSA or ovotransferrin. These mixtures were then diluted 5-fold over approximately 1

min with subsequent additions of 10 pi aliquots of 50 mM Na2HP04, pH 7.0, to give a final volume of 100 pi. Both assays were performed in separate wells of microtitre plates and the final amount of protein precipitation in each case was measured as turbidity at 360 nm with a Spectramax 250 microplate reader (Molecular Devices,

Sunnyvale, CA, USA).

97 4.2.4 NMR Spectroscopic Analysis of Interactions Between Clusterin and Reduced a-

Lactalbumin - Real-time one-dimensional ]H NMR spectra of samples of a-lactalbumin

(at 3.0 mg/ml) with or without added clusterin (0.7 mg/ml) were acquired at 500 MHz on a Varian Inova-500 NMR spectrometer. All samples were prepared in 570 pi of buffer

containing 50 mM Na2HP04, 0.1 M NaCl, 10% D20, pH 7.2. A series of spectra were acquired over time after the addition of 30 pi of a 0.4 M solution of deuterated DTT to each of the samples to give a final concentration of 20 mM. In each case, the acquisition of spectra was continued until no further changes in the one-dimensional spectrum were observed. In total, a sweep width of 5500 Hz was used over 9000 data points with an acquisition time of 0.82 s per scan and a delay between scans of 1.0 s. Sixteen scans were acquired per spectrum which, along with two blank control scans, gave a total time between successive spectra of approximately 33 s. An exponential line broadening of 3

Hz was applied to all NMR spectra prior to Fourier transformation.

98 4.3 RESULTS

4.3.1 Temperature Has Negligible Effect on the Oligomerisation State of Clusterin.

To examine the effect of increased temperature on the oligomerisation state of

clusterin, samples of clusterin were applied to a calibrated Pharmacia Superose-6 size

exclusion chromatography column immediately after a 1 hour incubation at either 37

°C or 50 °C and elution from the column was monitored as described in section 4.2.1.

Figure 4.3.1 shows that, under the conditions used, there was no significant difference

in the elution profiles for clusterin samples that had been heated at either 37 °C or 50

°C. This indicates that the oligomerisation state of clusterin remains relatively stable

over the temperature range of 37 - 50 °C.

1.2 1.2

a > 1

•S 0.8 0.8 o 0.6 - 0.6 < H 0.4 0.4

0.2 0.2 0* 0 jtS-.^-ta**-: 0 30 40 50 60 70 80 30 40 50 60 70 80

Time (min)

Figure 4.3.1: Effects of temperature on the oligomerisation state of clusterin. Samples of clusterin were analysed by size-exclusion chromatography at 50 °C (Panel A) and 37 °C (Panel B) as described in Methods (see 4.2.1). The absorbance at 280 nm (thick solid lines) were normalised and deconvoluted into peaks eluting at -71 min (thickdashed line), -63 min (thin solid line), -56 min (dotted line) and -40 min (dashed line) as described in Methods. The arrows show the elution positions of molecular weight markers which were (from left to right): Blue Dextran (2 MDa), catalase (232 kDa), aldolase (158 kDa), albumin (67 kDa) and ovalbumin (43 kDa). The results shown are representative of three independent experiments.

99 4.3.2 Elevated Temperature Slightly Enhances Substrate Binding by Clusterin.

In ELISA, the level of binding of purified clusterin to plate-bound stressed

BSA, GST or lysozyme was only slightly enhanced at elevated temperature (Figure

4.3.2). In experiments in which BSA or GST were used as the target proteins, the amount of substrate binding by clusterin was dependent on the temperature used

(Figure 4.3.2 A & B). Of the temperatures tested, clusterin binding was greatest at 50

°C, with progressively less binding occurring at 45, 42, and 37 °C. The binding of clusterin to plate-bound stressed lysozyme was similarly affected by elevated temperature. However, unlike the results for BSA and GST, clusterin bound most strongly with stressed lysozyme at 45 °C, with slightly less binding at 50, 42 and 37 °C

(Figure 4.3.2 C). The cause of this observed difference is not known. Nevertheless, the ELISA results demonstrate that elevated temperature only slightly enhances the binding of clusterin to stressed proteins.

4.3.3 The Chaperone Action of Clusterin is only Slightly Enhanced at Elevated

Temperature.

The effect of elevated temperature on clusterin's ability to inhibit stress-induced precipitation of insulin, BSA, a-lactalbumin and lysozyme was investigated and the results presented in Figure 4.3.3 A-D, respectively. Reduction of the test proteins in the absence of clusterin produced extensive precipitation. In each case, the rate of precipitation was more rapid when the target protein was reduced at temperatures greater than 37 °C, but the final amount of protein precipitation was similar regardless of the temperature used. In the presence of exogenous clusterin and at 37 °C, the final level of target protein precipitation at the end of each time course was significantly reduced due to the chaperone action of clusterin. When the target proteins were

100 stressed at 45 or 50 °C in the presence of clusterin, the final level of protein

precipitation was slightly less than that achieved at 37 °C, indicating that the chaperone

action of clusterin was only slightly enhanced at elevated temperatures.

8 02468 10 02468 10 c mO u 37 °C © o .A < 42 °C • 45 °C o 50 °C o 8 10 • Clusterin (jig/ml)

Figure 4.3.2: Results of ELISA measuring the temperature-dependent binding of clusterin to stressed proteins; (A) DTT-reduced bovine serum albumin (BSA), (B) Heat-stressed glutathione-S-transferase (GST), and (C) DTT-reduced Lysozyme (lys). Individual data points are the mean of triplicate measurements. In each case, the error bars shown represent standard deviations of the mean and in many cases are too small to be visible. The results shown are representative of two independent experiments.

101 0 50 100 150 0 100 200 300 <

100 200 300 100 200 300

Time (min)

Key O 37 °C no addition 37 °C + clusterin A 45 °C no addition 45 °C + clusterin • 50 °C no addition 50 °C + clusterin

Figure 4.3.3: Effects of elevated temperature on clusterin's chaperone activity. (A) Insulin (2 mg/ml) was reduced with 2 mM DTT at 37 °C, 45 °C, or 50 °C in the presence or absence of clusterin (400 Mg/ml); see key. (B) BSA (750 Mg/ml) was reduced with 20 mM DTT at 37 °C or 50 °C in the presence or absence of clusterin (300 Mg/ml). (C) a-lactalbumin (750 Mg/ml) was reduced with 20 mM DTT at 37 Tor 50 °C in the presence or absence of clusterin (300 Mg/ml). (D) Lysozyme (250 Mg/ml) was reduced with 20 mM DTT at 37 °C or 50 °C in the presence or absence of clusterin (700 Mg/ml). Each experiment was performed a minimum of three times and the individual traces shown are representative.

102 4.3.4 Low pH Enhances Binding of Clusterin to Stressed Proteins.

In chapter 3, clusterin was shown by ELISA to bind preferentially to stressed forms of ovotransferrin and lysozyme at near-physiological pH (i.e. pH 7.0). Here, the effects of mildly acidic pH on the binding of purified clusterin to stressed forms of ovotransferrin, GST, a-lactalbumin or BSA were investigated by ELISA (Figure

4.3.4). In all cases, regardless of whether the substrate protein was heat stressed or chemically reduced, the level of binding of purified clusterin to these substrates was clearly enhanced under mildly acidic conditions. Purified clusterin exhibited greatest binding to stressed proteins at pH 6.0, with progressively less binding occurring at pH

6.5 and 7.5. Similar results were obtained in ELISA experiments in which (i) GST, ovotransferrin or lysozyme were used as target proteins, and (ii) clusterin, in the form of pH-adjusted unfractionated human serum, was substituted for purified clusterin

(Figure 4.3.5). However, the level of increased binding of clusterin to stressed proteins at low pH was considerably greater for purified clusterin than for clusterin in unfractionated serum (compare Figure 4.3.4 to Figure 4.3.5). The most likely explanation for this observation is that serum proteins, undergoing partial unfolding at pH 6.0, were competing with plate-bound stressed protein for clusterin binding. At pH

5.0 and 4.0, the level of binding of purified clusterin to stressed target proteins was similar to that obtained at pH 6.0 (data not shown). Control experiments verified that pH had no effect on the amount of stressed proteins immobilised on the ELISA wells

(data not shown).

103 w s C3 -E o as

[clusterin] (jug/ml)

Figure 4.3.4: Effects of mildly acidic pH on the binding of clusterin to heat-stressed (A) ovotransferrin (ovo), (B) glutathione-5-transferase (GST), and reduced (C) ovotransferrin (ovo + DTT), (D) a- lactalbumin (a-lact) and (E) bovine serum albumin (BSA + DTT). All experiments were conducted in

50 mM Na2HP04 buffer at pH 6.0 ( • ), pH 6.5 ( • ), or pH 7.5 ( • ) - see key. Data points shown represent triplicate measurements. In each case, the error bars shown represent standard deviations of the mean and in many cases are too small to be visible. Results shown are representative of three independent experiments.

104 1.5 /"J.

I 1.0 1 08 f 0.5 as 0.0 0 20 40 60 80 100 % serum

Figure 4.3.5: Effects of mildly acidic pH on the binding of clusterin (in unfractionated human serum) to heat-stressed GST at pH 6.0 (• ), pH 6.5 (• ), and pH 7.5 ( 9 ). Data points shown represent triplicate measurements. In each case, the error bars shown represent standard deviations of the mean and in many cases are too small to be visible. These results are representative of two independent e xperimen ts.

4.3.5 Low pH Enhances Cluster in-Mediated Inhibition of Protein Precipitation.

Results shown in Figures 4.3.4 and 4.3.5 clearly indicated that the binding of clusterin to stressed proteins in ELISA is enhanced at mildly acidic pH (see section

4.3.4). However, the effect of mildly acidic pH on the chaperone action of clusterin was unknown. To address this issue, a number of target proteins, either purified proteins or proteins present in unfractionated human serum, were stressed (i) in the presence or absence of clusterin and (ii) at pH ranging from 6.0 - 7.5. The results of these experiments are shown in Figure 4.3.6. For all target proteins tested in the absence of clusterin, the rate and extent of protein precipitation were similar regardless of the pH used. In the presence of exogenous clusterin, the extent of protein precipitation was reduced in a pH-dependent manner. For example, at pH 6.0, clusterin was able to suppress approximately 97% of ovotransferrin precipitation compared to

105 71% at pH 6.5 and 33% at pH 7.5 (Figure 4.3.6 A). A similar trend was observed for

GST and catalase (Figure 4.3.6 B & C). The corresponding levels of clusterin-

mediated inhibition of protein precipitation in unfractionated human serum were 83%

at pH 6.0 and 53% at pH 7.5 (Figure 4.3.6 D).

Figure 4.3.6: Effects of mildly acidic pH on clusterin's chaperone action. (A) ovotransferrin (Ovo), (B) GST, (C) catalase (Cat) or (D) proteins in unfractionated human serum (diluted 1 in 10), were stressed as described in Methods (section 4.2). Protein precipitation at pH 7.5, 6.5 or 6.0 and in the presence or absence of exogenous clusterin (see key) was measured as absorbance at 360 nmas a function of time. Clusterin-depleted serum (CDS) was prepared from normal human serum (NHS) by immunoaffinity chromatography, as described in Chapter 2; General Materials and Methods. Each experiment was performed a minimum of three times and the individual traces shown are representative.

106 Further decreasing the pH to 5.0 and 4.0 only marginally increased the ability of clusterin to inhibit heat-induced precipitation of proteins (data not shown). Taken together, the results presented here clearly show that mildly acidic pH enhances the ability of clusterin to inhibit the precipitation of stressed proteins. In contrast, mildly acidic pH had an opposite effect on the chaperone action of the cytosolic sHsp, a- crystallin. Figure 4.3.7 demonstrates that, at pH 6.0, a-crystallin was unable to inhibit the heat-induced precipitation of ovotransferrin whereas, at pH 6.5, 7.0 and 7.5, the final levels of protein precipitation were reduced by approximately 10, 76, and 88%, respectively. Similar results were obtained when catalase was used as a stressed protein target (data not shown).

Key

A -a-crys, pH7.5

A + a-crys, pH7.5

• + a-crys, pH7.0

• + a-crys, pH6.5 • + a-crys, pH6.0

Figure 4.3.7: Plot showing the effects of low pH on the chaperone activity of a-crystallin. Ovotransferrin (1 mg/ml) was heat-stressed in the presence (solid symbols) or absence (empty symbols) of a-crystallin at pH 7.5 (triangles), pH 7.0 (circles), pH 6.5 (squares) or pH 6.0 (diamonds) - see Key. For clarity, results in the absence of a-crystallin at pH 6.0, 6.5 and 7.0 are not shown. Each experiment was performed a minimum of three times and the individual traces shown are representative.

107 4.3.6 Clusterin Does Not Inhibit the Rapid Precipitation of Lysozyme or y-Crystallin.

Figure 4.3.8 shows that clusterin, at a subunit molar ratio of 1.0:20 clusterimy- crystallin or 1.6:1.0 clusterimlysozyme, is able to suppress the slow precipitation of y- crystallin undergoing heat stress at 60 °C and lysozyme undergoing DTT-mediated reduction at 42 °C by approximately 60% and 50%, respectively (Figure 4.3.8 A & B).

When either of these proteins is unfolded in 6M GuHCl and then diluted into phosphate buffer (as described in 4.2.3), they rapidly precipitate from solution. Using the same subunit molar ratio of clusterin: target as in the experiments measuring slow precipitation, clusterin was unable to suppress the rapid precipitation of either y- crystallin or lysozyme (Figure 4.3.8 C & D). Even when the clusterin:target ratios were doubled, clusterin failed to inhibit the rapid precipitation of either protein (data not shown). Control proteins, BSA and ovotransferrin, were also unable to inhibit the rapid precipitation of lysozyme and y-crystallin (Figure 4.3.8 A & B).

108 Y - Crys Lys

o

0 10 20 30 40 0 100 200 300 Time (min) Time (min)

• No Additions A + Clusterin

1.2 0.8 TJ 1.0 c 0 0.6- o 0.8 NO 0.4- <*1 0.6 0.4 0.2- 0.2 o-L- 0 K II • No Additions Q +BSA 1 *—I—^—• •—i—• •—: • + Clusterin • + Ovotransferrin

Figure 4.3.8: Effect of clusterin on the slow and rapid precipitation of y-crystallin and lysozyme. Slow precipitation: (A) y-crystallin and (B) lysozyme was stressed by heat and reduction, respectively, as described in Methods (4.2) and protein precipitation was measured as absorbance at 360 nm as a function of time. These experiments were performed either in the absence of clusterin (empty symbols) or in the presence of clusterin (solid symbols) (see key). The data points shown represent the mean of triplicate measurements. Rapid precipitation: (C) y-crystallin and (D) lysozyme were denatured in 6 M

GuHCl and rapidly precipitated following 10- and 5-fold dilution with 50 mM Na2HP04 buffer, pH 7.0, respectively. These experiment were performed in the absence of clusterin (empty bar) or in the presence of clusterin (solid bar), BSA (shaded bar), or ovotransferrin (striped bar). The results shown in C and D are end point measurements and the error bars shown are standard errors of the mean of triplicate measurements. None of the proteins added had a significant effect on the extent of rapid precipitation of either y-crystallin or lysozyme (/5>0.05, r-test). The results shown are representative of three independent experiments.

109 4.3.7 Real-time ]HNMR Analysis of the Interaction Between Clusterin and Reduced a-

Lactalbumin.

The ability of clusterin to interact with and confer stability to chemically reduced a-lactalbumin has previously been established (Humphreys et al, 1999). To investigate this interaction at the molecular level and to determine the conformation of reduced a-lactalbumin with which clusterin specifically interacts, the time-dependent reduction of a-lactalbumin, in the presence or absence of clusterin, was monitored in real-time using a 500 MHz NMR spectrometer, as described in section 4.2.4. Figure

4.3.9A & B shows the aromatic region of selected one-dimensional !H NMR spectra of solutions of a-lactalbumin, either non-reduced or reduced with DTT, in the absence

(Figure 4.3.9A) or in the presence (Figure 4.3.9B) of clusterin. The !H NMR spectra for non-reduced a-lactalbumin, alone or with clusterin, are characterised by well- dispersed and highly defined resonance peaks (at 0 s; Figure 4.3.9A & B). Similarities between these spectra indicate that the dominant resonances arose from a-lactalbumin

(as expected since clusterin was present at a much lower level in the mixture).

Therefore, changes in resonance intensity or dispersion in subsequent spectra were solely attributable to conformational changes undergone by a-lactalbumin.

As can be seen from Figure 4.3.9A, in the absence of clusterin, incubation of a- lactalbumin with DTT resulted in a rapid alteration in the NMR spectra. This involved transition from (i) the resolved NMR spectrum of the native species (at 0 s) to (ii) a broad spectrum with some dispersion, reflecting the presence of partially unfolded intermediates containing mostly secondary and some tertiary structure (Ii, e.g. at 384 s), to (iii) an even broader spectrum lacking any dispersion, produced by an

intermediate with some secondary structure but no tertiary structure (I2, e.g. at 639 s).

With time (e.g. 2170 s), the I2 state of a-lactalbumin was completely lost as the

110 reduced protein aggregated and precipitated from solution. Upon removal of the

sample from the spectrometer after the experiment, a substantial amount of protein

precipitate was present in the NMR tube.

B

NW~>WV»WM^. 2160 s

•.v~ 1450 s

0s (No DTT)

100 £ 80 -4->*> c 40 > •S3 20

Ill Similar spectra were recorded for a-lactalbumin undergoing DTT-mediated reduction in the presence of clusterin (Figure 4.3.9B). However, in this case, the

progression from the native species (at 0 s) to the I2 state of a-lactalbumin was slower

(compare Figure 4.3.9A & B). In addition, whereas the I2 state of a-lactalbumin was abolished by 2170 s when the protein was reduced in the absence of clusterin, in the

presence of clusterin the I2 state of reduced a-lactalbumin was still present 5400 s after

the addition of DTT (Figure 4.3.9C). Thus, clusterin had a stabilising effect on the I2 state of a-lactalbumin. When the sample was removed from the NMR spectrometer after the experiment, the protein solution had remained clear, indicating that clusterin had successfully formed complexes with partly unfolded a-lactalbumin, inhibiting its precipitation from solution.

Figure 4.3.9C shows a plot of the intensity of resonance (at 6.8 ppm) versus time after the addition of DTT for a-lactalbumin in the presence or absence of clusterin. The resonance at 6.8 ppm arises from the (3,5) aromatic ring protons of tyrosine residues of reduced a-lactalbumin and is relatively isolated from the broad envelope of other aromatic resonance. Since native a-lactalbumin does not have a significant resonance at this chemical shift, monitoring its intensity with time after the

addition of DTT provided a means of following the build-up and decay of the I2 state without complications from resonances arising from the native state of a-lactalbumin.

From Figure 4.3.9C, it can be seen that for a-lactalbumin reduced either in the presence or absence of clusterin, the intensity of resonance at 6.8 ppm increased rapidly to a maximum at approximately 380 s, indicating that clusterin had no

significant effect on the build-up of signal from the I2 state. In addition, since the

maximum NMR resonance intensity from the I2 state of a-lactalbumin occurs when all

112 four of its disulfide bonds are reduced (Carver et al, 2002), the data presented in

Figure 4.3.9C indicate that clusterin does not significantly affect the rate of disulfide bond reduction in a-lactalbumin.

After maximum resonance intensity was reached, the signal progressively

decayed as reduced a-lactalbumin (in the I2 state) either (i) aggregated in the absence of clusterin, or (ii) complexed with clusterin (Figure 4.3.9C). As a means to compare the rates of decay of the 6.8 ppm resonance for a-lactalbumin, the equation for the first-order process was fitted to the data shown in these figures. As a result, the calculated first-order rate constants for the time-dependent resonance intensity decay were (1.81 ± 0.09) x 10"3 s"1, in the absence of clusterin, and (5.07 ± 0.13) x 10"4 s"1, in the presence of clusterin. Therefore, under the conditions described, clusterin

approximately tripled the lifetime of the I2 state of a-lactalbumin.

Recently, a similar pattern of binding by clusterin to insulin has also been recorded using *H NMR spectroscopy (data known shown). Together with the results showing the binding of clusterin to a-lactalbumin, this result suggests that clusterin may specifically interact with the disordered intermediately folded molten globule states of stressed proteins.

113 4.4 DISCUSSION

A large number of publications are devoted to the analysis and characterisation of the structure and chaperone action of sHsps under both physiological and stress conditions (e.g. exposure to elevated temperature). These studies indicate that interactions between sHsps and stressed proteins are achieved through hydrophobic interactions. This may also be true for clusterin. Cross-linking of the hydrophobic probe bis(l-anilinonaphthalene-8-sulfonate) (bis-ANS) to clusterin reduces the latter's ability to inhibit the precipitation of its substrates by competing with stressed proteins for hydrophobic regions on clusterin (Poon et al, 2002; see Appendix). Many studies indicate that under physiological conditions sHsps exist in solution as large multimeric complexes. Upon exposure to heat stress, many sHsps undergo substantial structural changes that may result in (i) loss of native-like motifs, leading to the formation of molten globule-like structures, (ii) dissociation of the multimer to form oligomers and other lower-order forms, and (iii) increased protein surface hydrophobicity, which can lead to an enhancement of chaperone activity (section 4.1). Taken together, these observations suggest that, for these sHsps, the non-aggregated, lower-order forms of the proteins are responsible for their chaperone action.

Similarly, clusterin has also been shown to exist in solution as multimeric structures. However, unlike the sHsps, results shown in Figure 4.3.1 demonstrate that elevated temperature does not significantly affect the oligomerisation state of clusterin.

In addition, exposure of clusterin to temperatures between 37 °C and 50 °C did not substantially alter its ability to interact with heat-stressed or chemically reduced target proteins (Figure 4.3.2), or its ability to inhibit precipitation of proteins undergoing

DTT-mediated reduction (Figure 4.3.3). Furthermore, as reported in Poon et al, 2002, a decrease in the amount of exposed hydrophobicity on clusterin was detected with

114 increasing temperature. Taken together, these results demonstrate that, at elevated temperatures, clusterin remains oligomerised, has decreased exposed hydrophobicity, but is still capable of functioning as a chaperone. This suggests that the hydrophobic regions of clusterin that become less exposed to solution at elevated temperatures are probably discrete from those regions that associate during clusterin oligomerisation and that may form parts of the chaperone-binding site(s).

The effects of mildly acidic pH on the oligomerisation state of clusterin have previously been documented (Hochgrebe et al, 2000). Gel filtration studies indicated that, at physiological pH (i.e. pH 7.5), clusterin exists as aggregates of varying numbers of associated afS heterodimers, which dissociates into monomers at mildly acidic pH (pH 5.5 - 6.0). In contrast to results obtained at elevated temperatures, dissociation of the clusterin oligomers occurs at low pH and is accompanied by increased exposure of hydrophobic regions on clusterin and increased binding of clusterin to native protein and polysaccharide ligands (Hochgrebe et al, 2000;

Pankhurst et al, 1998). However, mildly acidic pH induces very little change in the overall secondary structure of clusterin as detected by CD analysis. Therefore, it was reasoned that under these conditions, the enhanced binding of clusterin to the various native target ligands could not be attributed to any hydrophobic regions that would have only been exposed to solvent as a result of substantial structural alterations.

Hydrophobic regions normally buried in the clusterin-clusterin interfaces but which become exposed to solvent following pH-dependent dissociation of clusterin oligomers may be responsible for the enhanced ligand binding.

Results presented in this chapter indicate that at mildly acidic pH, the chaperone activity of clusterin is significantly enhanced (Figure 4.3.4 - 4.3.6). It is possible that hydrophobic regions responsible for the enhanced binding of clusterin to

115 native ligands at low pH may also contribute to the enhanced interaction between clusterin and stressed proteins under these conditions. In a recent report, it was suggested that clusterin contains three regions of natively disordered or molten globule-like structures that are sensitive to protease digestion (Bailey et al, 2001). The binding of the hydrophobic probe, ANS, to clusterin protected at least one of these disordered regions from trypsin-mediated proteolysis, indicating that it is hydrophobic and may therefore contribute to the binding site(s) on clusterin for stressed protein ligands.

The effect of low pH on the structure of clusterin is strikingly similar to that of increased temperature on the sHsps. There are no previously published studies of the effect of low pH on the chaperone action of sHsps, although results shown in Figure

4.3.7 indicate that the in vitro chaperone action of the sHsp, a-crystallin, is dramatically reduced when exposed to mildly acidic conditions. The enhanced ligand- binding and chaperone action of clusterin at low pH may have important physiological relevance. A phenomenon known as acidosis occurs at sites of tissue damage or inflammation where the local pH can drop below 6. Acidosis has been reported to occur at sites of inflammation, cardiac ischaemia, infracted brain, and in the brains of

Alzheimer's sufferers (McGeer and McGeer, 1997). Under these conditions, clusterin oligomers may dissociate and the enhanced binding/chaperone actions of the heterodimeric species could help to inhibit the aggregation and deposition of inflammatory and/or toxic proteins which would otherwise exacerbate pathology.

Under stress conditions, the unfolding of a target protein occurs via a series of partly structured intermediates or molten globule states (see Figure 4.1.1). Depending on the amount of native-like structure, molten globules can be classified as being

116 ordered (Ii) or disordered (I2). Of these, the I2 intermediates are the least stable because they are less structured; they contain some secondary structure but very little, if any, tertiary structure, and hence expose significantly greater amounts of hydrophobicity to solution than the Ij intermediates. This increased exposure of

hydrophobicity causes the I2 molten globules to slowly aggregate and precipitate via the irreversible off-folding pathway.

Small Hsps such as a-crystallin have been shown to interact specifically with

the less stable I2 form of molten globules. Small Hsps have no affinity for target proteins that are either in the native state or in the stable Ii molten globule state

(Lindner et al, 1997). Similar results have been presented in Poon et al, 2002; the report showed that clusterin does not bind to (i) native a-lactalbumin, (ii) its stable, monomeric, partly structured intermediates, or (iii) the fully unfolded form of a-

lactalbumin. Clusterin binds only to the reduced form of a-lactalbumin, which is I2- like (on the basis of CD spectroscopic analysis; Okazaki et al, 1994) and is known to slowly aggregate and precipitate (Humphreys et al, 1999; Lindner et al, 1997). The

ability of clusterin to specifically interact with the I2 form of a-lactalbumin suggests that kinetic factors might be important in regulating the chaperone action of clusterin.

To examine this hypothesis, precipitation experiments were conducted to test the ability of clusterin to inhibit (i) the slow precipitation of y-crystallin and lysozyme

(induced by heat and reduction, respectively), versus (ii) the rapid precipitation of the same proteins induced by dilution from denaturant. Results presented in Figure 4.3.8 demonstrate that clusterin was effective in suppressing the slow precipitation of y- crystallin and lysozyme, but was unable to prevent these same target proteins from rapid precipitation. Taken together, these results indicate that, like sHsps, clusterin

117 binds specifically to slowly aggregating proteins on the off-folding pathway and that kinetic factors are important in its chaperone action.

Real-time !H NMR spectroscopic analysis of the interaction between clusterin and a-lactalbumin suggests that this process occurs in two phases. A similar two-step binding pattern has also been described for the interaction of the sHsp, a-crystallin, upon interaction with reduced a-lactalbumin (Lindner et al, 1997; Lindner et al,

2001). Clusterin's ability to extend the lifetime of the I2 state of a-lactalbumin about

3-fold (Figure 4.3.9C) indicates that, during the first phase, the non-aggregated

(probably monomeric) form of the intermediate was stabilised through an interaction

with clusterin. The progressive loss of the NMR spectrum of the I2 state of a- lactalbumin during reduction in the presence of clusterin indicates that, during the second phase, larger clusterin-a-lactalbumin complexes form, producing species of large molecular mass that would tumble slowly and give rise to very broad NMR spectra. When chemically reduced with DTT, a-lactalbumin is known to form HMW complexes with clusterin (Humphreys et al, 1999).

In conclusion, the data presented in this chapter indicate that the chaperone action of clusterin is dependent on both conformational and kinetic factors. Exposure to mildly acidic conditions induces dissociation of clusterin oligomers, resulting in an increased exposure of hydrophobic regions to solution, and an enhancement of clusterin's chaperone action. Like sHsps, clusterin cannot inhibit very rapidly precipitating proteins. Instead, clusterin preferentially binds to partly unfolded, slowly aggregating protein intermediates. Taken together, these results suggest that clusterin is a pH-dependent chaperone that functions by specifically interacting with disordered

118 molten globule states of slowly aggregating proteins on the irreversible off-folding

pathway.

A subset of the results presented in this chapter have been published in the following journal papers:

Poon, S., Treweek, T.M., Wilson, M.R., Easterbrook-Smith, S.B., and Carver, J.A. (2002) "Clusterin is an extracellular chaperone that specifically interacts with slowly aggregating proteins on their off-folding pathway" FEBS Letters: 513, 259-266. (see Appendix)

Poon, S., Rybchyn, M.S., Easterbrook-Smith, S.B., Carver, J.A., Pankhurst, G.J., and Wilson, M.R. (2002) "Mildly Acidic pH Activates the Extracellular Molecular Chaperone Clusterin" J. Biol. Chem: 277, 39532-39540. (see Appendix)

119 Chapter 5

Production and Expression of Vectors Encoding Wild Type and Mutant Clusterin Molecules

120 5.1 INTRODUCTION

Clusterin is a highly conserved protein that is constitutively expressed and secreted by a wide range of cell types (section 7 J.2) (Rosenberg and Silkensen, 1995;

Jenne and Tschopp, 1992). Under normal conditions, clusterin is primarily expressed in epithelial cells and at fluid-tissue interfaces where it has been suggested to protect cell membranes from the harmful effects of prolonged exposure to "biologically active fluids" such as bile, gastric juice, pancreatic juice and urine (Jordan-Starck et al, 1994;

Aronow et al, 1993). Expression of clusterin is up-regulated by cellular stress (e.g. heat shock, oxidative stress) (Clark and Griswold, 1997) as well as a variety of signals related to the growth state and oncogenic status of the cell. Clusterin gene expression has also been detected in a number of diseases such as Alzheimer's disease (May et al,

1989; 1990), scrapie (Duguid et al, 1989), epilepsy, various human gliomas (Danik et al, 1991), and in the retinas of patients with retinitis pigmentosa (Jones et al, 1992).

The role played by clusterin in these and other diseases is not known. However, in vitro results presented in (Humphreys et. al, 1999) and in this thesis indicate that clusterin is a chaperone; it preferentially interacts with and stabilises proteins that are partially unfolded, as was shown when clusterin suppressed the precipitation of chemically-reduced proteins in unfractionated human serum (section 3.3.2). Therefore, it is reasonable to suggest that clusterin might also function in vivo to stabilise non- native protein structures and to confer cellular protection against a wide range of stress conditions.

The region(s) of clusterin structure responsible for the chaperone action of the protein have not been identified. However, sequence analysis of the primary structure

121 of clusterin has revealed several regions that could be of functional importance. These include five regions of predicted amphipathic a-helices (Bailey et al, 2001), a motif analogous to the cytolytic components of the complement membrane-attack complex

(Jenne and Tschopp, 1989; 1992), a putative nucleotide-binding site (Tsuruta et al,

1990) and two predicted myosin-like coiled-coil domains (Jordan-Strack et al, 1992)

(Figure 5.1.1). According to Bailey et al, 2001, the five regions of predicted amphipathic a-helices correspond to residues 19-39, 149-169, 217-234, 309-327, and

407-422 of the mature clusterin molecule. Using the coils algorithm, the two coiled- coil a-helical regions were predicted to correspond to residues 18-77 and 296-328 of mature clusterin (Wilson and Easterbrook-Smith, 2000). The a-helices are of particular interest for elucidating the regions of clusterin that are important for its chaperone action. Since amphipathic a-helices are thought to be important in mediating interactions with hydrophobic molecules, they may be responsible for the propensity of clusterin to interact stably with lipids and hydrophobic domains present in other proteins. In addition, they may also explain why clusterin could (i) mediate cell-cell interactions, (ii) bind with high affinity to a wide range of biological ligands including amyloid (3 (1-40), complement components, and lipids such as those found in high-density lipoproteins, and (iii) function in vitro as a chaperone protein, binding to denatured proteins to prevent aggregation and precipitation.

Several studies by others working on clusterin have been conducted to identify structural regions responsible for the promiscuous binding ability of this protein. In one particular study, 34 peptides, each consisting of 15 amino acids, were synthesised from hydrophilic regions of human clusterin and their ability to promote cell adhesion in vitro was assessed (Silkensen et al, 1999). Of these, two peptides, designated Clu-

29 (corresponding to residues 337-351) and Clu-32 (residues 381-395), completely

122 inhibited clusterin-mediated cell adhesion, whereas the peptide, Clu-10 (residues 96-

110), promoted cell adhesion These results suggest that the region of human clusterin corresponding to peptide Clu-10 may be important for cell adhesion. In another study, sequence analysis predicted three regions of natively disordered, molten globule-like structures containing putative amphipathic a-helices (Bailey et al, 2001). Limited digestion with trypsin revealed 11 trypsin cleavage sites that were located within or immediately proximal to the predicted molten globule-like domains, indicating that these regions are highly flexible. ANS binding to these domains resulted in the protection of one trypsin cleavage site, suggesting that at least one of the predicted molten globule-like regions is hydrophobic and is likely to be involved in the binding of protein or lipid substrates by clusterin. A more recent study tested whether the binding of clusterin to ligands could be inhibited by competitive binding with other clusterin ligands or by anti-clusterin monoclonal antibodies (Lakins et al, 2002;

Appendix). The results of this study suggested that clusterin has three separate binding sites for: stressed proteins, unstressed ligands, and the cell surface receptor low-density lipoprotein receptor-related protein-2 (LRP-2). In the same report, the ability of purified human serum clusterin (/isClus) to suppress stress-induced protein precipitation was compared to that of purified recombinant Pichia pastoris - derived clusterin (rpClus), which has variable proteolytic truncations of the C-terminal regions of the a-chain and the N-terminal region of the f3-chain. Results showed that, even with the truncations, r/?Clus had similar in vitro chaperone activity to hsClxxs.

Therefore, collectively, these studies strongly suggest that the site(s) responsible for the chaperone action of clusterin is(are) likely to be located more towards the N-terminal region of the a-chain and the C-terminal region of the P-chain. However, the exact

123 location of the binding sites for stressed proteins, unstressed ligands, and LRP-2 remains to be identified.

As an initial step towards identifying the structural region(s) responsible for the promiscuous binding ability and chaperone action of human clusterin, expression plasmids encoding wild type or a range of mutant human clusterin molecules were developed. Mutations (in the form of proline-substitutions or truncations) were introduced into specific regions of the clusterin sequence (Figure 5.1.1) that were thought likely to be of functional importance (e.g. predicted amphipathic or coiled-coil a-helices, and regions towards the N-terminus of the a-chain and C-terminus of the (3- chain). By inducing significant conformational changes or truncations to these regions and testing for their effects on the binding and chaperone abilities of clusterin, the functional regions of the protein may be determined.

Table 5.1.1 lists characteristics of the various clusterin mutants developed in this study. Proline-substitution mutants clusA27P, clusQ155P, clusM228P, clusL322P, and clusA413P were developed by substituting helix-disrupting proline for various amino acids in the clusterin sequence. For these mutants, the sites of proline- substitution were determined from GOR IV secondary structure analyses of the predicted regions of a-helix in human clusterin (section 5.2.7). These sites were specifically chosen because, at these residue positions, proline was predicted to be most disruptive to the helical structures (which are possible functional domains of clusterin; see above). Truncation mutants clusT17 and clusT78 were developed by removing portions of the a-chain N-terminus whilst mutants clus294T, clus356T and clus413T had portions of the p-chain C-terminal tail removed. Expression of wild type and mutant clusterin genes was attempted in Spodoptera frugiperda IPLB-Sf9 (Sf9) insect cells (Invitrogen, CA, USA), a clonal isolate derived from the parental

124 Spodoptera frugiperda cell line IPLB-Sf-21-AE (SCI) of the ovaries of fall armyworms. Sf9 insect cells were chosen (above other expression systems; e.g., mammalian, yeast, bacteria) as the host for the production of recombinant wild type and mutant clusterin due largely to their reported ability to express recombinant proteins at very high levels (up to 25-50% of the total cell protein; Luckow and

Summers, 1988). In addition, Sf9 have also been reported to carry out posttranslational modifications (e.g. fatty acid acylation, glycosylation, and phosphorylation) in a manner similar to that of mammalian cells (Vialard et al, 1995). This chapter describes (i) processes involved in the construction of expression clones for the production of recombinant wild type and mutant human clusterin, and (ii) results of expression of wild type and mutant clusterin molecules in Sf9 insect cells.

125 c/: OJ C/D

•S3 c Si 8 S c 8 o -a XI 2 xc> OJ 'o T3 o c m -a si a3 O C -a O Q, CJ o £ c X 3 o CO Xo) C >> H ^ T3 T3 o cj O CJ o T3 •a 3 •p T3 OJ B W5 OJ p a II u

ha

Q_ I 8 00 * o C\J<(>>

>- •o c B 8 -< 13 =3

c Z5 jQ (/) CQ.

^ 5 e«

E 73 a .5 c o cd o 'B c c x Si C o CJ .S o I cd .5 .5 o •s a, i Q. Q. 8 x; cd cd Si P 2 C+- cj X E 5 CJ o cj I .S3 -o Cfl CO. 3 C/0 •5 •a on. CO. •a •a x 3 x x •ss x a £ • S CU CU 13 .Su x: x; e cu x: cu .S S cu c > > > u cj q o > a. a. a. a. a, a. fci fci *i CU cu I cu •a 23 c cd I c C C C C 3 c o O O o O 1 . ,_,. w -t—1 J_J 13 o 3 3 3 3 3 3 .^2 .^i o: cfl O .X XI XI Xl Xl I 3 3 3 3 3 c c c c o o o o CU OJ

CU o e c cd P-, 0- 00 H H H 1 m (N

127 5.2 METHODS

5.2.1 Determining the Sites for Proline Substitution or Truncation in Human Clusterin

- The sites for proline-substitution in human clusterin was determined using the GOR

IV secondary structure prediction program (located at http://npsa-pbil.ibcp.fr/cgi- bin/npsa_au tomat.pl?page=npsa_gor4.html). Briefly, the program utilises predictive algorithms developed by Gamier, Gibrat, and Robson (Gamier et al, 1996) to predict whether an amino acid in a given protein (in this case, clusterin) is part of an a-helix,

(3-strand, turn, or coil. The GOR TV algorithm is based on the theoretical propensity of amino acid residues to form particular secondary structures based on each residue's side-chain structure and its interaction with the structure of neighbouring residues.

According to this algorithm, if an amino acid is flanked by residues that prefer to be a part of an a-helix, it is likely that the same amino acid also forms part of the helix, even though its individual helix preference may be low. When a polypeptide sequence is entered into the program, two outputs are generated; the first output assigns each amino acid residue within the sequence to one conformational state of a-helix (H), extended or beta strand (E), or coil (C); the second gives the probability values for each secondary structure at each amino acid position.

To determine the optimal site for proline-substitution in human clusterin, sequences corresponding to each of the predicted regions of amphipathic or coiled-coil a-helices of clusterin were entered into the GOR TV program. These regions were chosen for site-directed mutagenesis because they are likely to be involved in clusterin's ligand-binding ability or chaperone action (section 5.1). Systematic substitution of amino acids within these sequences for proline was performed and the total number of non-helix-forming residues was determined. For each case, the residue position at which proline caused the greatest disruption of helix structure (determined

128 by the number of predicted non-helix-forming residues) was chosen as the optimal site for proline-substitution.

As the a-chain N-terminus and (3-chain C-terminus of clusterin have been predicted to contain functional domains, possibly involving the amphipathic or coiled- coil a-helices (section 5.1), all of the truncation sites were placed at various residue positions within these regions of the protein. A total of five truncation sites were chosen; these include residue positions 17 and 78 of the a-chain, and positions 294,

356, and 403 of the [3-chain (section 5.3.1).

5.2.2 Quickchange™ Site-Directed Mutagenesis (Overview) - A total of five proline- substitution mutants were developed in this study (Figure 5.1.1). The substitution of helix-disrupting proline for various amino acids in the clusterin sequence was achieved using the Quickchange™ site-directed mutagenesis kit (Stratagene, CA, USA). An overview of the procedures involved is summarised in Figure 5.2.1. Basically, it utilises a supercoiled double-stranded DNA (dsDNA) vector (in this case, pTZ-HT7;

Figure 5.2.2A) with the full-length clusterin gene and two synthetic oligonucleotide primers that (i) contain the desired mutation (i.e. a codon which encodes for the amino acid, proline) and (ii) are complementary to opposite strands of the vector. The oligonucleotide primers are extended during temperature cycling by Pfu Turbo DNA polymerase, during which a mutated circular plasmid containing staggered nicks is generated. Staggered nicks occur due to the inability of the two ends of each DNA strand of the mutated plasmid to anneal together following temperature cycling (Figure

5.2.1). Dpn I (a restriction endonuclease which is specific for the 5'-GmATC-3' sequence of methylated and hemimethylated DNA) is then added to digest unreacted

129 parental DNA templates and to select for mutation-containing synthesised DNA. Since the template DNA is isolated from DNA adenine methylase positive (dam+) E. coli and is therefore dam methylated, it is susceptible to Dpn I digestion. The nicked mutated plasmid is then transformed into XL 1-Blue supercompetent E.coli cells where they are repaired.

1. Gene in plasmid with (Q) target site ( • ) for V—V_^V* mutation. 1 2. Denature plasmid and anneal fr^ primers ( $ ) containing I\^>lL A)r desired mutations ( • ).

—1 *- — v

3. Extend and incorporate 1 mutagenic primers resulting (' ( J? V in nicked circular strands. ** — — — Trie locations where these nicks occur are indicated 1 by(X). 4. Digest the methylated, non-mutated fry) parent DNA template with Dpnl. Vr\—-V^

1 5. Transform into XL1-Blue ^^ Supercompetent cells.

Figure 5.2.1: Overview of the Quickchange™ site-directed mutagenesis method as outlined instruction manual which accompanied the kit (Stratagene).

130 In this study, the full-length clusterin gene in the plasmid, pTZ-HT7 (Figure

5.2.2A) was mutated using PCR to produce gene constructs encoding the various proline-substitution mutants. The mutagenic oligonucleotide primers used were clusA27P (GTCAATAAGGAAATTCAAAATCCCGTCAACGGGGTGAAACAGA), clusQ155P (CACATGCTGGATGTCATGCCCGACCACTTCAGCCGCGCG), clusM

228P (TTCCAGCCCTTCCTTGAGCCCATACACGAGGCTCAGCAG), clusL322P

(TTGACCAGGAAATACAACGAGCCCCTAAAGTCCTACCAGTGG), and A413P

(CCTAAATTTATGGAGACCGTGCCCGAGAAAGAGCTGCAGGAA). The mutated codons that were introduced using these primers are underlined. For each primer listed, the complementary primer was also used. All primers were synthesised by

Invitrogen/Life Technologies, NY, USA. The compositions of the PCR mixtures are indicated in Table 5.2.1. PCR was performed in a MJ Research Model PTC-200 DNA

Engine thermocycler (MJ Research Inc., NY, USA) using the conditions indicated in

Table 5.2.2.

After PCR amplification, the mutant construct plasmids were subsequently treated with 10 units of Dpn I restriction enzyme (Invitrogen/Life Technologies, NY,

USA). Thereafter, all Quickchange™ - mutated plasmids were transformed into XL1-

Blue supercompetent cells as specified below (section 5.2.2) and allowed to propagate.

131 A. pTZ-HT7 fl (IG) (~ 4250 kb) bla (Apr)

Clus (-1.3 kb)

rep (pMBl)

B.pDONR™201 T1-T2 attP1 (4470 kb) ccdB

attP2

attR1 cmr C. pXJNsect-Dest38 Actin (12419 kb) Promote

HR3 Actin polyA

Figure 5.2.2: Map of various plasmid vectors that were used to create the Gateway™ expression clones, as described in 5.2.4. A. pTZ-HT7; a plasmid derived frompTZ19R vector (MBI Fenrentas)containing the clusterin gene. B. pDONR™201 (Invitrogen); upon insertion of the gene of interest during the BP recombination reaction, this plasmid becomes the entry clone. C. pXINsect-Dest38 (Invitrogen); during the LR reaction, recombination of this plasmid with the entry clone results in the formation of the expression clones.

132 Control Sample

Reaction Reactions

Composition Amount Used Amount Used

10X Reaction Buffer 5 pi 5 pi

pWhitescript 4.5-kb Control Plasmid (5 ng/pl) 10 ng -

Double-Stranded Template DNA (pTZ-HT7) - 50 ng

Olionucleotide Control Primer #1 125 ng -

Olionucleotide Control Primer #2 125 ng -

Mutagenic Oligonucleotide Primer #1 - 125 ng

Mutagenic Oligonucleotide Primer #2 - 125 ng

dNTPMix(lOmM) lpl lpl

PfuTurbo DNA Polymerase (2.5 U/pl) lpl lpl

Double-Distilled Water To 50 pi To 50 pi

Table 5.2.1: List and amounts of various components used for Quickchange mutagenesis of the clusterin gene in the plasmid, pTZ-HT7.

Cycles Temperature Time

Denaturation 30 seconds 1 95 °C Denaturation 12 95 °C 30 seconds

Annealing 1 minute 55 °C Elongation 2 minutes/kb of plasmid length 68 °C

Table 5.2.2: Cycling parameters for the PCR-based incorporation of proline residues into clusterin cDNA via Quickchange™ mutagenesis.

133 5.2.3 Transforming into XLl-Blue Supercompetent Cells - XL 1-Blue supercompetent cells were thawed on ice for 1-2 minutes. For each reaction mixture, 50 pi of XLl-

Blue cells were dispensed into a pre-chilled Falcon 2059 polypropylene tube. To these,

1 pi of Dpn I-treated DNA from each reaction was added, gently mixed and incubated on ice for 30 minutes. All transformation reactions were heat pulsed for 45 seconds at

42 °C and then placed immediately back on ice for 2 minutes. Thereafter, 0.5 ml of

NZY+ broth (1% (w/v) NZ amine (casein hydrolysate), 0.5% (w/v) yeast extract, 0.5%

(w/v) NaCl, 12.5 mM MgCl2, 12.5 mM MgS04, 20 mM glucose, pH 7.5) preheated to

42 °C was added to the mixtures and incubated whilst shaking at 240 rpm for 1 h at 37

°C. 250 pi of each transformation mixture was then spread onto LB agar plates containing 50 pg/ml ampicillin, 80 pg/ml 5-bromo-4-chloro-3-indoyl-P-D- galactopyranoside (X-gal) and 20 mM isopropyl-l-thio-(3-D-galactopyranoside (IPTG) and the cells were allowed to propagate overnight (12-16 h) at 37 °C. Thereafter, cells that had been successfully transformed with the mutated plasmid were selected by blue-white colour screening with X-gal and IPTG. These cells were picked for small- scale culture (25 ml) from which the mutated plasmid was extracted and purified using the Wizard® Plus miniprep DNA Purification System (Promega, WI, USA)- refer to section 5.2.5. All constructs were verified by DNA agarose gel electrophoresis.

5.2.4 Gateway™ Cloning Technology (Overview) - The Gateway™ cloning system involves three separate procedures; (i) generation of a^B-flanked PCR products, (ii) creation of entry clones via the BP reaction, and (iii) creation of expression clones via the LR reaction. An overview of these procedures is shown in Figure 5.2.3. The benefit of using this system is that clusterin cDNA can be transferred into an entry vector, from which it can then be transferred via a simple one-step reaction into one of

134 many different Gateway -compatible expression vectors (Figure 5.2.4). Vectors developed for this system contains unique regions (referred to as att sites) that flank the gene of interest (i.e. clusterin) and allow for site-specific and directional recombination cloning and subcloning of the gene without the need for restriction enzymes and ligase.

In this study, the Gateway cloning system was used to clone and subclone wild type and mutant clusterin sequences into the appropriate donor (pDONR™201; Figure

5.2.2B) and expression (pXTNsect-DEST38; Figure 5.2.2C) vectors. The expression of both wild type and mutated recombinant clusterin was achieved using Sf9 insect cells.

5.2.4.1 Generation of attB-Flanked PCR Products - As mentioned above, the first step in the Gateway™ cloning of the gene for wild type and mutant clusterin is the generation of PCR products that are flanked by the attB sites. The PCR products for developing wild type and proline-substitution mutants were created via two-stage PCR amplification using two sets of overlapping clusterin wild-type primers. The first set of primers contained: (forward primer) - (i) a 12 bp segment of the attBl sequence

(AAAAAGCAGGCT), (ii) a Kozac sequence (GCCACC), (iv) a start codon (ATG), and (v) a 15 bp template specific sequence; (reverse primer, complementary to the coding sequence) - (i) a 12 bp segment of the attBl sequence (AGAAAGCTGGGT),

(ii) a stop codon (TCA), and (iii) a template-specific sequence. The second set of primers (adapter-primers attBl and attBl) were used to install the complete attBl

(GGGGACAAGTTTGTACAAAAAAGCAGGCT) and attBl (GGGGACCACTTTG

TACAAGAAAGCTGGGT) sequences (Figure 5.2.5).

135 Vector containing gene of interest PCR (e.g. clusterin) Amplificattioi n ^ clus

OKrrt r clus att P2 att B1 att B2 ccdB PCR Product Donor Vector (pDONR-201) BP Reaction I ccdB attU^^'attll clus attm att ^2 By-product Entry Clone V

/ •

attmXs^. y?'attR2 O" ccdB Destination Vector clus (pXINsect-Dest38) LR _ Entry Clone Reaction I

affPITv ^7attP2 O" ccdB By-product att Br^i_gJf att B2 clus IExpression Clone

Transformation into target cells (e.g. Sf-9 Insect Cells)

Figure 5.2.3: General overview of the Gateway™ cloning system. This system which was used to clone and subclone wild-type and mutated clusterin genes, involves three main steps. In the first, recombination (attB) sites are introduced to the flanks of the genes of interest via PCR amplification. In the second step, these genes are transferred by recombination into a donor plasmid (pDONR™201) to create the entry clones. During the last step, the genes are transferred via recombination into a destination vector (e.g. pXINsect-Dest38) to produce the expression clones. The resulting expression clones are subsequently transformed into target cells (e.g. Sf9) which then express the proteins of interest.

136 t3 c 2 •5 .2 .S 22 e3 £ -o >i cd M B .23 O

.53 — .s c/3 w 3 U o O g o 00 to o o J3 JS ^ 00 |=-. „ u •S 2 2 c •B 5 O o in * > S B 2 .S 03 •S .S&CD C a B P ° 5 « h c o C« (1) O5 V> ^ o i cj o a. c

137 04 T—I oo a) gj cu T3 OH B HE -B J= CJ o oo •3 •" c 03 o CC u a u u 03 C5 CS ^ u u C3 *-* < CJ OQ C3 o u C« K) U < H £3 to 3i u CD a O O tN 0< <; 3 CJ u •a ^ CN CJ H u EH 2 H o< EH oo oo H EH EH Q , > o u CJ ' 03 •o 0) u u o 1 1 § C E-" OH J CD Q QC „ U u u H •3 B •a < X < < u y a, X H T3 X a CJ X < ~-^ X X X « o en tn I X X X o •" 1) O P X X X •s ^ X X X 00 fa X B S X CD CO X CD G X X X a C E X Q. X X 00 CD CO o X X X S CD CD X c X B P 3CD X! X X X c X 0H CJ = si CT x X 2 & X CD CD X ^" CD X X U OH CD CJ X m CO X o ^^ X X X OH <*- J2 S X X X X X e B •" -a X X X •o ° o X X X X X 5 o oo P X X X X X t 8 &> X X X B '5 .5 a. X 2. a. o g co X a X I WD X C>D TOE " O u < X ? CJ y o3 ^ CD < u u cs aj CD > o EH u EH ,, . sig e< *H b O a a ft CT CL CD Q a > B H u CD MH EH < CB CD O O u U « .eg 1- T3 o C5 o U o > ^ 52 as CD u o CJ CB ^^ CN JS o CU o CD B u CD B u ^ * O a 3 H 3 -a -c CD 0X1 H H H o a EH u CJ CE si cd -£ feb,a W5 <-> ^ WD EH 3 C TO ^3 OH t_> C8 *-> cs en g 5 8 >" CD -*- o e CD CJ en CO .. B, 00 aa S03 T3 en M UH >>) f en O mM O

138 The PCR products for developing wild type and proline-substitution mutants were amplified from pTZ-HT7 and Quickchange™-mutated plasmids, respectively.

The primers used during the first round of PCR were clusWT12bForward

(AAAAAGCAGGCTCGGCCACCATGATGAAGACTCTGCTG) and clusWT12b

Reverse (AGAAAGCTGGGTCTCACTCCTCCCGGTGCTT). PCR products for the truncation mutants were created using truncation-specific primers via a two-step PCR amplification similar to that used to amplify the full-length clusterin gene from pTZ-

HT7. The truncation-specific primers used in the first stage of PCR were clusT1712bForward (AAAAAGCAGGCTCGGCCACCATGAGTAAGTACG TCA

AT), clusT7812bForward (AAAAAGCAGGCTCGGCCACCATGGGAGTGTGCA

TTGAG), clus294T12bReverse (AGAAAGCTGGGTCTCAGTTGGTGGAACAG

TCCAC), clus356T12bReverse (AGAAAGCTGGGTCTCAGCCTTGCGTGAGGTT

TGC), and clus403T12bReverse (AGAAAGCTGGGTCTCACCTGGAGACTTCTA

CAGG). In cases where a forward truncation-specific primer was used, the clusWT12bReverse primer was also used. The clusWT12bForward primer was used when the reverse truncation-specific primers were used. All PCR amplifications were performed in an MJ Research Model PTC-200 DNA Engine thermocycler using the conditions indicated in Table 5.2.3. All PCR products were verified by agarose gel electrophoresis (section 2.3) prior to their transfer into the donor vector, pDONR™201 to create the Entry clones.

139 Stages Cycles Temperature Time

1 95 °C 30 seconds

Denaturation 30 seconds 95 °C 1 Annealing 10 30 seconds 45 °C Elongation 2 minutes/kb of plasmid length 68 °C Denaturation 30 seconds 95 °C Annealing 2 20 30 seconds 55 °C Elongation 2 minutes/kb of plasmid length 68 °C

Table 5.2.3: Cycling parameters for PCR amplification of the clusterin gene flanked by attB sites.

5.2.4.2 Creating Entry Clones Via the BP Reaction - The transfer of the PCR products into pDONR-201 was conducted via the BP reaction; so-called due to the conversion of the attB sequence to attL, producing a by-product with an attP sequence (see Figure

5.2.3). The reaction mixtures consisted of the components listed in Table 5.2.4.

140 Negative Positive Sample Control Control Tubel Tube 2 TubeX

PCR Product (100 fmol) — — 1-10 |il

pEXP7-tet Positive Control (50 ng/|il) — 2ul

pDONR201 Vector, 150 ng/|il 2 |il 2 |il 2 |il

BP Reaction Buffer (5X) 4ul 4 |il 4ol

TE Buffer 10 uT 8ul To 16 |il

Table 5.2.4: Amounts of various components used to create entry clones via the BP recombination reaction. pExp7-tet is a 1.4 kb linear DNA which encodes the tetracycline resistance gene (Tcr) and its promoter for expression. When reacted with the donor vector, it can be used to estimate kanamycin resistant (Km') transformants that contain the transferred Tcr gene.

To each of the sample reactions, 4 JLXI of BP CLONASE Enzyme mix (Invitrogen) was added, vortexed briefly (2 s) twice, and incubated at 25 °C for 60 min. Thereafter,

2 \i\ of Proteinase K solution (2 mg/ml) was added and further incubation conducted at

37 °C for 10 min. 1 u\l of each sample reaction was transformed into 50 JLLI of LIBRARY

EFFICIENCY DH5a competent cells prior to incubation on ice for 30 min. All samples were heat-shocked at 42 °C for 30 s, placed on ice for a further 1-2 min, and diluted by the addition of 450 |il S.O.C. medium (2% (w/v) tryptone, 0.5% (w/v) yeast extract,

0.05% (w/v) NaCl, 20 mM glucose; dissolved in de-ionised water, pH 7.0). 100 ul of transformed cells were subsequently plated onto Luria-Bertani (LB) agar (1% (w/v) tryptone, 0.5% (w/v) yeast extract, 2% agar, 1% NaCl; dissolved in de-ionised water, pH 7.0) plates containing 50 ug/ml kanamycin and incubated overnight at 37 °C. All construct plasmids were isolated and purified using the Wizard® Plus Miniprep DNA

Purification kit (section 5.2.6).

141 5.2.4.3 Creating Expression Clones Via the LR Reaction - The creation of the expression clones was achieved via the LR reaction; so-called due to the process involving the reaction of an Entry clone (attL) with a Destination clone (attR), creating a new Expression clone (attB) (see Figure 5.1.4). The destination vector used was pXINsect-Dest38. The required reaction mixtures consisted of the following:

Negative Positive Sample Control Control Tubel Tube 2 TubeX

Entry clone (require 300 ng) — — 1-11 Ml

pENTR-gus Positive control (50 ng/|Ltl) — 2 (xl —

Destination vector (require 300 ng) 1-11 pi 1-11 jil 1-11 |il

LR Reaction Buffer (5X) 4 |il 4 |il 4 |il

TE Buffer lOul 8 |il To 16 |il

Table 5.2.5: Components and their amounts used to create expression clones for wild type and mutant clusterin. pENTR-gus is a 1.8 kb plasmid encoding the gus gene. When reacted with a destination vector, the resulting expression clone contains both E. coli and eukaryotic translational signals upstream of the gus gene, allowing for native expression in E. coli, yeast, insect, and mammalian cells depending on the destination vector used.

To each of the sample reactions (shown above), 4 |itl of LR CLONASE Enzyme mix

(Invitrogen) was added, vortexed briefly (2 s) twice, and incubated at 25 °C for 60 min.

Thereafter, 2 |xl of Proteinase K solution (2 mg/ml) was added and further incubation conducted at 37 °C for 10 min. 1 \il of each sample reaction was transformed into 50

|il of LIBRARY EFFICIENCY DH5a competent cells as above. 100 ul of transformed cells were plated onto LB plates containing 100 Ug/ml ampicillin and allowed to propagate overnight at 37 °C. All constructed plasmids were isolated using the

142 Wizard" Plus Miniprep DNA Purification kit (section 5.2.5) and verified by automated, bi-directional DNA sequencing for their correct sequence prior to transfection into insect (Sf9) cells. DNA sequencing was performed by Tim Gleeson (Canadian

Molecular Research Services, ON, Canada).

5.2.5 Transfection of Sf9 Insect Cells - Into separate wells of a 24-well plate, each containing 1.5 ml Sf-900 II serum-free medium (SFM), 9 x 105 Sf9 cells were added and allowed to attach by incubating overnight at 27 °C. 1 ug of sterile, purified expression plasmid DNA was diluted to a final volume of 200 |il in Sf-900 II SFM before being added to 200 ul SF-900 E SFM containing 12 ul CELLFECTIN® reagent and 0.1 ug pBmAmeo, a pUC-based plasmid which confers neomycin resistance to transfected cells. The resulting mixture was then incubated at room temperature for 45 min. During this incubation, the Sf9 cells were washed once with 2 ml Sf-900 II SFM after which, 800 ul of SFM was added to the wells. To these, 200 ul of the DNA/

CELLFECTIN complexes was added after the 45 minutes incubation. Thereafter, the plates were incubated for a further 5 h at 27 °C, washed once with 2 ml of Sf-900 II

SFM, replaced with 1 ml of SFM, and then returned to the 27 °C incubator. After 48 h, the transfected Sf9 cells were resuspended in SFM supplemented with 5% (v/v) foetal calf serum (FCS) and selected with 500 ug/ml GENETICIN®. Transfected insect cells were maintained in the presence of GENETICIN® for 2 weeks before tissue-culture supernatants of these cells were screened for the presence of clusterin.

5.2.6 Wizard® Plus Miniprep DNA Purification System - The small-scale isolation and

(R) purification of plasmid DNA from bacterial cultures was achieved using the Wizard

Plus miniprep DNA purification kit (Promega, WI, USA). Following the specified

143 protocol, 10 ml of bacterial culture was centrifuged at 1,400 g for 10 min, resuspended in 400 |ll of cell resuspension solution, and transferred to a 1.5 ml microcentrifuge tube. To this, 400 ul of cell lysis solution was added and the tube inverted four times to ensure adequate lysis of the cells. 400 ul of neutralisation solution was immediately added and mixing was achieved by inverting the tube several times. The lysate was then cleared by centrifugation at 10,000 x g for 5 min. The supernatant, containing the

DNA, was subsequently passed through a minicolumn, which was connected to a 3 ml disposable syringe containing 1 ml of Wizard® Minipreps DNA Purification resin. To each syringe, 2 ml of column wash buffer was added and gently pushed through with a syringe plunger. The minicolumn was then detached from the syringe, transferred to a

1.5 ml microcentrifuge tube and centrifuged at 10,000 x g for 2 min. Finally, the minicolumn, to which 50 ul of nuclease-free water was added, was transferred to a new microcentrifuge tube and centrifuged at 10,000 x g for 20 s to elute the plasmid DNA.

Purified DNA was stored at -20 °C.

144 5.3 RESULTS

5.3.1 Chosen Sites for Proline-Substitution and Truncation in Human Clusterin

The sites for proline-substitution in the helical regions of clusterin were determined using the GOR IV secondary structure prediction program as described in section 2.2.1. The results are summarised as follow:

Amphipathic helix 1 (19-39) - Six single proline substitution mutants were tested, including E23P, I24P, A27P, V28P, V31P, and K32P (data not shown). Of these,

A27P was predicted to be the most disruptive to helical structure, causing a contiguous run of thirteen non-helix-forming residues at 19-31 (i.e., 65% disruption). The native sequence is predicted to have only one non-helix-forming residue at 30.

Coiled-coil 1 (40-77) - This predicted coiled-coil region has a high propensity to form a helix. Even with three proline substitutions clustered close together (at residue positions 49, 53, and 58), only partial disruption of the long helical region is predicted

(a stretch of thirteen non-helical residues predicted; i.e., 35% disruption). Therefore, it was decided that site-directed mutation is not the way to study this particular structural feature.

Amphipathic helix 2 (149-168) - Six single proline substitutions were tested within this region, including V153P, M154P, Q155P, D156P, H157P, and F158P (data not shown). M154P, Q155P and H157P were found to be equally the most disruptive, each causing a run of eight contiguous non-helix-forming residues at 154-161 (i.e., 42% disruption). Within 149-168, the native sequence has only 1/20 residues that are

145 predicted to be non-helix forming. Q155P was chosen (above M174P) because the former is likely to also reduce the amphipathicity of the predicted helix [138].

Amphipathic helix 3 (217-234) - Of the eleven single proline substitutions tested, including H219P, A220P, M221P, F222P, Q223P, F225P, L226P, E227P, M228P,

I229P, and H230P (data not shown), M228P was predicted to be the most disruptive to

helical structure, causing a contiguous run of eleven non-helix-forming residues from

217-227 (i.e., 50% disruption).

Amphipathic helix 4 (309-327) - Of the eleven single proline substitution mutations

analysed, which include V311P, E313P, L315P, T316P, R317P, K318P, Y319P,

E321P, L322P, L323P, and K324P (data not shown), L322P and K324P were

maximally disruptive, each associated with a prediction of a contiguous run of six non-

helix-forming residues at 319-324 and 320-325, respectively (i.e., 33% disruption); the

native sequence has no non-helix-forming residues within 309-327). L322P was

chosen as the site for proline substitution simply because it is slightly closer to the

centre of the predicted helix.

Coiled-coil 2 (296-308) - Of six single-site proline substitution mutants analysed,

including K300P, L301P, R302P, E304P, L305P, and D306P (data not shown), the

most disruptive to coil structure were L301P and R302P, both of which were predicted

to produce a run of six contiguous non-helix-forming residues (in both cases at 296-

301). However, even the native sequence was predicted to have a run of three

contiguous non-helix-forming residues at 296-299. So, these mutants were not

predicted to be very "good" helix-breakers.

146 Amphipathic helix 5 (407-423) - Of the ten single proline substitution mutants analysed, which include E410P, T411P, V412P, A413P, E414P, K415P, A416P,

L417P, E419P, and Y420P (data not shown), A413P was predicted to be the most disruptive to helix. A contiguous stretch of six non-helix-forming residues was predicted at 407-412 (i.e., 37.5% disruption).

In summary, the optimal sites for proline substitution as determined from GOR IV analysis of the helical regions of human clusterin are positions 27 (A27P), 155 (Q155P),

228 (M228P), 322 (L322P), and 413 (A413P).

Five truncation sites within clusterin were designed to selectively delete portions of the a-chain N-terminus and (3-chain C-terminus - the region of clusterin proposed to contain the functional motifs of the protein (section 5.1). These sites included residue positions T17, T78, 295T, 356T, and 403T. In the case of T17, the truncation site (which occur in the a-chain) was placed close to the start of the first predicted a-helix to cause the deletion of the first 16 (non-helix-forming) amino acid residues of the mature clusterin molecule. The truncation site for T78, also located in the a-chain, was chosen to cause the deletion of the first 77 amino acids of the clusterin molecule, including the first predicted amphipathic (residue 19-39) and coiled-coil (residue 40-77) a-helices.

The truncation site at residue 295 was selected to delete 132 amino acid residues from the (3-chain C-terminus, a region of the protein which contains the fourth and fifth predicted amphipathic a-helices (residues 309-327 and 407-423) as well as the second coiled-coil a-helix (residue 296-308). Truncation sites at residues 356 and 403 were designed to cause the deletion of 71 and 24 residues from the (3-chain C-terminus, respectively. In both cases, the resulting truncated clusterin molecules should contain the

147 all amphipathic and coiled-coil a-helices, with the exception of the fifth amphipathic

5.3.2 Site-Directed Mutagenesis

The substitution of various amino acid residues for the helix-disrupting proline in clusterin was achieved using the commercially available Quickchange™ site- directed mutagenesis kit as outlined in section 5.2.2. A total of five mutant plasmid constructs were developed using this kit (Table 5.1.1), all of which were subsequently transformed into XLl-Blue E. coli cells. Successfully transformed cells containing the mutated plasmids were selected by blue-white colour screening with X-gal and IPTG.

Visual scoring of these cells indicated very high (>95%) mutagenesis efficiencies.

Agarose gel electrophoresis confirmed the sizes of mutant recombinant clusterin amplicons (data not shown).

5.3.3 PCR Amplification of Wild Type and Mutant Clusterin Genes and Creation of Entry

Clones. cDNA encoding wild type and proline mutants were cloned using Gateway- specific attB-flanked primers via two-step PCR amplification (section 5.2.3.1). While wild type clusterin cDNA was cloned from the pTZ-HT7 plasmid, cDNA encoding the proline mutants were cloned from the various Quickchange™ - mutated plasmids. PCR products for the truncation mutants were developed using truncation-specific primers via a similar two-step PCR amplification process (section 5.2.3.1). The resulting attB- flanked PCR products were analysed on a 1% agarose gel (Figure 5.3.1) prior to their insertion into separate donor (pDONR™) plasmids via BP recombination reactions

(section 5.2.3.2) to create the entry clones. A total of eleven entry clones were developed

(one containing wild-type clusterin cDNA, five containing Quickchange - mutated

148 cDNA, and five containing truncated cDNA). All entry clones, in the form of uncut plasmids, were analysed by agarose gel electrophoresis (Figure 5.3.2) prior to transformation into LIBRARY EFFICIENCY DH5a cells, which were selected with kanamycin and allowed to propagate.

Lanes 12 3 4 5 6 7 8 9 10 11 12 13 14

Figure 5.3.1: Image of an ethidium bromide-stained 1% agarose gel showing attB-PCR products; Lane 1, DNA marker (1 kb DNA Step Ladder; Promega, WI, USA); Lane 2, attB-clusWT (wild type); Lane 3, a«B-clusT17; Lane 4, affB-clusT78; Lane 5, a«B-clus294T; Lane 6, affB-clus356T; Lane 7, attB- clus403T; Lane 8, DNA marker; Lane 9, aftB-clusWT; Lane 10, a«B-clusA27P; Lane 11, attB- clusQ155P; Lane 12, a«B-clusM228P; Lane 13, affB-clusL322P; Lane 14, a?fB-clusA413P.

149 A Lanes B Lanes 12 3 4 5 6 7 12 3 4 5 6 7 10

5 ^-m-s 4 linear

Figure 5.3.2: Image of an ethidium bromide-stained 1 % agarose gel showing entry clones for; A) wild type and truncation mutants; Lane 1, DNA marker (1 kb DNA Step Ladder; Promega, WI, USA); Lane

2, pENTR-clusWT (wild type); Lane 3, PENTR-clusT17; Lane 4, PENTR-clusT78; Lane 5, pENTR- clus294T; Lane 6, pENTR-cIus356T; Lane 7, pENTR-clus403T. B) wild type and proline-substitution mutants; Lane 1, DNA marker; Lane 2, pENTR-clusWT; Lane 3, pENTR-clusA27P; Lane 4, pENTR- clusQ155P; Lane 5, pENTR-clusM228P; Lane 6, pENTR-clusL322P; Lane 7, pENTR-clusA413P. The major bands shown represent entry clones in the supercoiled state. Minor bands represent nicked and linearised plasmids

5.3.4 Creation and Analysis of Expression Clones.

The expression clones were developed via LR recombination reactions between the entry clones and the destination vector, pXINsect-Dest38 (section 5.2.3.3). All expression clones were subsequently analysed by agarose gel electrophoresis (Figure

5.3.3) and then transformed into LIBRARY EFFICIENCY DH5a cells. Successfully transformed cells containing the expression clones were selected with ampicillin.

Automated bi-directional DNA sequencing (section 5.2.3.3) confirmed that each of the expression clones contained the correct gene sequence (Appendix).

150 A Lanes B Lanes 1234567 1234567

Figure 5.3.3: Image of an ethidium bromide-stained 1% agarose gel showing expression vectors for; A) wild type and truncation mutants; Lane 1, DNA marker (1 kb DNA Step Ladder; Promega, WI, USA); Lane 2, pXINsect-Dest38-clusWT (wild type); Lane 3, pXINsect-Dest38-clusT17; Lane 4, pXINsect-Dest38-clusT78; Lane 5, pXINsect-Dest38-clus294T; Lane 6, pXINsect-Dest38-clus356T; Lane 7, pXINsect-Dest38-clus403T; B) wild type and proline-substitution mutants; Lane 1, DNA marker; Lane 2, pXTNsect-Dest38-clusWT; Lane 3, pXINsect-Dest38-clusA27P; Lane 4, pXINsect- Dest38-clusQ155P; Lane 5, pXINsect-Dest38-clusM228P; Lane 6, pXINsect-Dest38-clusL322P; Lane 7, pXINsect-Dest38-clusA413P.

5.3.5 Expression of Wild-type and Mutant Clusterin in Sf-9 Cells.

Expression plasmids containing cDNA for wild type or mutant clusterin molecules were transiently co-transfected with the pBmA:neo plasmid into Sf9 cells with CELLFECTIN®. The selective agent GENETICIN® (500 ug/ml) was added to the growth medium (Sf-900 E SFM; supplemented with 5% (v/v) FCS) 48 hours after transfection. Successfully transfected cells grew very slowly. Two weeks after transfection, tissue-culture supernatants from individual wells containing transfected cells were collected and analysed by immuno-dot blot and western transfer procedures

(sections 2.4 & 2.7). Immuno-dot blot analysis of these tissue-culture supernatant detected the presence of an expressed protein (which was absent in non-expressing control cells) that was reactive with G7, 4ID, and 78E anti-clusterin antibodies (Figure

5.3.4) - indicating the presence of clusterin in each of the supernatants. Further

151 analysis of the tissue-culture supernatant fractions by western blotting also revealed the presence of clusterin (Figure 5.3.5). Taken together, these results indicated that (i) the

Sf9 cells were successfully transfected with the expression clones containing either wild type or mutated clusterin cDNA, and (ii) the truncations or single amino acid substitutions tested did not cause any aberrant conformational changes to the antibody recognition/binding sites (at least, for the antibodies tested).

serum Clusterin (positive control) non-trans feet S/9 wild-type • o supernatant clusA27P clusT17 • • clusQ155P • • clusT78 clusM228P clus294T • • clusL322P clus356T • •

clusA413P • • clus403T

Figure 5.3.4: Immuno-dot blot analysis of tissue culture supernatants obtained from transfected Sf9 cells two weeks after transfection. The presence of clusterin was detected using a cocktail of G7, 41D, and 78E anti-clusterin monoclonal antibodies (as described in section 2.7). A sample of purified human serum clusterin was also dotted onto the membrane to serve as a positive control. Non-transfected Sf9 cells did not appear to express detectable amount of clusterin (circle).

However, it appears that the transiently transfected Sf9 insect cells were unable to produce large amounts of recombinant clusterin. SuperSignal® WestPico western blot substrate, which can detect 10"12 grams of HRP-labelled proteins during enhanced chemiluminescence detection (ECL; section 2.6) failed to detect the presence of recombinant clusterin in any of the Sf9 supernatants, even when the supernatants were concentrated 20-fold. Instead, the detection of western blotted Sf9-derived clusterin from most Sf9 supernatants was successful only when SuperSignal® WestFemto, an

152 ECL substrate that can detect 10" grams of HRP-labelled proteins, was used (Figure

5.3.5).

Lanes

Figure 5.3.5: Western blot of a 10% SDS-PAGE gel of purified human serum and Sf9-derived recombinant clusterin. Lane 1, purified human serum clusterin; Lane 2, Sf9-derived wild type clusterin; Lane 3, clusA27P; Lane 4, clusQ155P; Lane 5, clusM228P; Lane 6, clusL322P, Lane 7, clusA413P; Lane 8, clusT17; Lane 9: clusT78; Lane 10, clus294T; Lane 11, clus356T; Lane 12, clus403T.

Furthermore, examination of the immunoblot reveals that the Sf9-expressed wild type and mutated clusterin had molecular masses that were approximately 15 kDa smaller than expected (Figure 5.3.5). For instance, wild type serum clusterin has a molecular mass of approximately 78-80 kDa (Figure 5.3.5, lane 1). However, recombinant wild type Sf9 clusterin was detected as a 64 kDa protein (Figure 5.3.5, lane 2), as was the proline- substitution mutant clusterin molecules (Figure 5.3.5; lanes

3-7). The truncation of clusT78 results in the removal of 77 amino acid residues from the N-terminus of the a-chain, which also includes the removal of a glycosylation site

(at residue N64) from the mature protein. As a result of this truncation, the estimated size of the resulting clusterin molecule, if it was expressed in mammalian cells, is -65 kDa). Since ClusT78 was detected as a 53 kDa protein on the Western blot (Figure

153 5.3.5; lane 9), it was also smaller than expected. The smaller than expected masses of the Sf9-expressed clusterin molecules may have resulted from one of three possibilities; the expressed proteins were partially glycosylated, incompletely translated or inappropriately cleaved by proteolytic enzymes. Incomplete translation resulting in the expression or production of truncated clusterin molecules is unlikely because DNA sequencing confirmed the correct complete sequence for the various wild type and site-directed mutant clusterin cDNAs in the expression vectors. In addition, since multiple bands (corresponding to truncated portions of clusterin) were not detected in any of the supernatant fractions, it would be unlikely that inappropriate proteolysis was responsible for producing the undersized clusterin molecules.

Therefore, accepting this reasoning, the most likely cause for the size discrepancy may be incomplete glycosylation of the mature Sf9-derived clusterin molecules. However, further experiments would be required to confirm this.

154 5.4 DISCUSSION

To understand the fundamental relationship of a protein's structure to its function, it is necessary to identify the amino acids and/or structural motifs that are involved in the activity of the protein. A common approach towards such an understanding involves comparing the amino acid sequences of proteins (isolated from different species) that have the same biological function. By comparing the amino acid sequences of such proteins, it may be possible to identify conserved regions of amino acids that are essential for the protein's function. Indeed, many sHsps have been studied in this way. Sequence analysis of sHsps have revealed that, despite being structurally divergent, they share a common sequence of amino acids (the "a-crystallin domain") which is essential for the chaperone action of these proteins (section 3.7).

Another approach in the study of a protein's structure-function relationship involves the determination of the three-dimensional structure of the protein. The interactions between many enzymes (such as lysozyme and DNA polymerases) and their substrates have been studied by initially determining the three-dimensional structure of the enzyme. Unfortunately, neither of these approaches is suitable for elucidating the functional domains of clusterin due to (i) a lack of sequence homology with other known proteins and (ii) the absence of a crystal structure.

The development in the last decade of genetic engineering techniques such as site-directed mutagenesis offers an alternative to the above approaches. The basic principle of site-directed mutagenesis involves the introduction of specific mutations into the coding sequence of a protein such that when the gene encoding the protein is expressed in a suitable host, specifically mutated proteins are produced. Screening of

155 these mutants for function can identify the structures that are responsible for the function(s) of the native protein.

This chapter describes the processes involved in the development of recombinant clusterin mutants - potential tools for the identification of structural elements involved in the chaperone action of human clusterin. Mutations in the form of proline-substitutions or truncations were introduced into the coding sequence of human clusterin (situated in the vector pTZ-HT7) using synthetic mutagenic oligonucleotide primers (sections 5.2.1 & 5.2.4.1). Next, the wild type and mutant cDNAs were incorporated into the donor vector (pDONR™201) via BP recombination reactions to produce corresponding entry clones (section 5.2.4.2). From these vectors, the cDNAs were subsequently transferred via LR recombination reactions into an appropriate destination vector (pXINsect-Dest38) to produce the expression clones

(section 5.2.4.3). Finally, these expression clones were transiently transfected into Sf9 insect cells where they were incorporated into the host cell genome from which the cDNAs were transcribed and translated to produce the various clusterin molecules

(section 5.2.5). Sf9 insect cells were chosen as the host cells for the production of recombinant clusterin due to their reported ability to produce high levels of properly post-translationally modified recombinant proteins.

Despite these reported abilities of Sf9 insect cells, results presented in this chapter indicate that transiently transfected Sf9 cells were unable to produce high levels of recombinant clusterin. Since the Sf9 transfectants were not cloned, the Sf9 cell cultures may have contained a large number of non-transformants/non-secretors that may have outnumbered secreting transformants. Therefore, even though the Sf9 cultures appeared to be viable and highly confluent, the population of secreting transformants may have been small; hence the low level of expressed recombinant.

156 Analysis of the expressed proteins by immunoblotting also revealed that Sf9-expressed wild type and proline-substituted mutant clusterin molecules were approximately 15 kDa smaller than purified serum clusterin (section 5.3.5; Figure 5.3.6). Since DNA sequencing had verified the correct full-length sequence for genes encoding wild type and mutant clusterin, it was highly unlikely that this size discrepancy resulted from truncations of the coding sequence. In addition, it was unlikely that inappropriate proteolysis was responsible for the size discrepancy since if it was the case, there should have been other bands (corresponding to proteolytic fragments of clusterin) detected. Therefore, the size discrepancy most likely resulted from reduced glycosylation of the expressed proteins. Little is known about the glycosylation mechanisms of Sf9 cells, although it is generally believed that Sf9 is capable of carrying out post-translational modifications (such as glycosylation) in much the same way as vertebrate cells (Vialard et al, 1995). Interestingly, a recent report has indicated that Sf9 insect cells have difficulties in adding the more complex sugars, such as galactose and sialic acid (Pfeifer, 1998). Since clusterin is known to contain sialyl

Lewis type oligosaccharides (Kapron et al, 1997), it is likely that Sf9 insect cells are incapable of expressing fully glycosylated clusterin.

The inability of Sf9 insect cells to express large amounts of correctly processed recombinant clusterin has prompted a reinvestigation into the possible use of other host systems for the expression of high levels of recombinant wild-type and mutant clusterin. At present, a number of other host systems are available for gene expression and recombinant protein production, including bacteria, filamentous fungi, plants, yeast and mammalian cells (reviewed in "The recombinant protein handbook: Protein purification and simple purification", Amersham pharmacia biotech, 2000).

Traditionally, bacterial expression systems have been the preferred choice due to (i) the

157 availability of a wide range of cloning vectors, (ii) an extensive literature, (iii) high protein yield obtained from moderate scale cultures, and (iv) economy. However, the key disadvantage of bacterial cells, which prevented their use in this study, is their inability to perform any post-translational modifications that may be necessary for the expression of properly structured and fully functional clusterin.

Expression of recombinant wild type clusterin has previously been attempted using the Pichia pastoris yeast expression system. Like mammalian cells, yeast cells are known to perform glycosylation and disulfide bond formation. In addition, the advantage of using yeast expression systems is that yeast can be used to express large amount of protein at low cost. However, the P. pastoris expression system is not suitable for the expression of recombinant wild-type or mutant clusterin due to inappropriate proteolysis of the C-terminus of the a-chain and the N-terminus of the (3- chain (Lakins et al, 2002).

Mammalian expression systems could be most appropriate for the expression of recombinant clusterin. One limitation to many mammalian systems (which prompted the use of Sf9 cells in this study) is the relatively low expression levels that can normally be achieved using these systems. However, the use of mammalian expression system will ensure that correct post-translational modifications to wild-type clusterin and its mutants will be obtained. Successful clusterin expression has been achieved in cultured porcine smooth muscle (Moulson and Millis, 1999; Millis et al,

2001), baby hamster kidney (BHK-21) fibroblast (Pilarsky et al, 1993), mink lung epithelial (Reddy et al, 1996), Shionogi mammalian carcinoma (Renny et al, 1994), and Madin-Darby canine kidney (MDCK) (Hartmann et al, 1991; Urban et al, 1987) cell lines. Preliminary results have shown that cultured Chinese Hamster Ovary

(CHO)-expressed wild type clusterin co-migrates with human serum clusterin,

158 indicating that, unlike Sf9-expressed clusterin, it was properly post-translationally modified (Fay Dawes; unpublished results). In addition, the concentration of expressed wild type clusterin in transfected CHO tissue culture supernatant was estimated to be around 10 ug/ml, significantly greater than that obtained with transfected Sf9 cells. In light of these results, future efforts will be directed towards the use of the CHO- expression system for the production of recombinant clusterin molecules. By utilising the versatility of the Gateway™ cloning system (see Figure 5.2.4), transfer of the mutated clusterin cDNAs will simply require a one step recombination reaction between the entry clones developed in this study and an appropriate Gateway™- specific mammalian expression vector.

In conclusion, clusterin cDNA was mutated by site-directed proline substitutions and selected truncations. Using the Gateway™ cloning technology developed by Invitrogen/Life Technologies, a series of expression clones containing wild-type or mutated clusterin cDNA were successfully constructed and transfected into Sf9 cells. However, the results obtained strongly suggest that Sf9 cells are incapable of producing fully glycosylated clusterin. To produce properly processed wild-type and mutated clusterin for future studies, a different expression system (e.g.

CHO) will certainly be required.

159 Chapter 6

Conclusions

160 A number of environmental conditions, collectively referred to as stress, can lead to protein deposition, unfolding and aggregation. Some of these adverse conditions include extremes of temperature or pH, oxidants, charged metal ions, and molecular crowding. All living organisms respond to these stress conditions by rapidly increasing the synthesis of a diverse group of proteins called molecular chaperones. The primary function of molecular chaperones is to recognise and bind unfolded or partially folded proteins. By doing so, they prevent illicit intermolecular interactions and hence protein aggregation and precipitation. Of the molecular chaperone superfamily, the best characterised chaperones belong to the stress-induced proteins that are collectively known as the heat shock proteins (Hsps). All heat shock proteins belong to one of five families: HsplOO, Hsp90, Hsp70, Hsp60 and the small heat shock protein (sHsp) family. Some Hsps (e.g. Hsp60) have ATPase activity and are involved in protein folding in vivo. Many others (e.g. sHsps) lack ATPase activity and function by preventing the aggregation of cellular proteins accumulating as a result of stress. In a recent study, clusterin was shown to specifically inhibit the stress- induced precipitation of a broad range of structurally divergent substrate proteins in vitro, thus acting like a molecular chaperone. The chaperone action of clusterin has been compared to that of the sHsp family (Humphreys et al, 1999).

Results presented throughout this thesis confirm the in vitro sHsp-like chaperone action of clusterin. Like sHsps, clusterin is capable of binding a wide range of structurally divergent substrate proteins. In addition to those investigated in

161 (Humphreys et al, 1999), clusterin was shown to inhibit (i) the heat- or reduction- induced precipitation of ovotransferrin, (ii) the precipitation of lysozyme induced by reduction, (iii) heat-induced precipitation of y-crystallin, and (iv) heat- and reduction- induced precipitation of proteins in diluted and undiluted human serum, respectively

(Figures 3.3.1 & 3.3.2). Using an approach similar to that described in (Humphreys et al, 1999), clusterin was shown to bind preferentially to stressed forms of ovotransferrin, lysozyme and a-lactalbumin to form solubilised high molecular weight complexes (Figures 3.3.3-3.3.5). Complex formation between the various stress- denatured proteins and clusterin suggests that stable binding of substrates is fundamental to clusterin's ability to inhibit stress-induced protein precipitation.

It is generally accepted that all chaperones (including sHsps) share the property of binding to exposed regions of hydrophobicity on partially unfolded, stressed conformations of proteins to prevent their irreversible aggregation. The interaction between clusterin and stressed proteins appears to occur in much the same way. This is supported by an experiment which showed that cross-linking of the hydrophobic probe, bis(l-anilinonaphthalene-8-sulfonate) (bis-ANS), to clusterin reduces the latter's ability to inhibit the precipitation of its substrates by competing with stressed proteins for hydrophobic regions on clusterin (Poon et al, 2001).

Small heat shock proteins differ from many other chaperones because (i) they carry out their chaperone action independently of ATP, and (ii) they have little or no inherent ability to refold stressed proteins. Here, results show that inhibition of protein precipitation by clusterin also occurs in an ATP-independent manner (Figure 3.3.1) and that ATP has no effect on the in vitro chaperone action of clusterin (Figure 3.3.6).

In addition, regardless of the presence of ATP, clusterin is unable to protect ADH or catalase from heat-induced loss of activity (Figure 3.3.7). Furthermore, clusterin,

162 alone or when complexed with a substrate protein, does not appear to have the ability to hydrolyse ATP (Figure 3.3.8). Taken together, these results strongly suggest that

ATP is not directly involved in the chaperone action of clusterin. These in vitro studies also suggests that clusterin is unlikely to possess the ability to refold partly unfolded proteins since most chaperones that are able to refold stabilised substrates require ATP to do so (lakob et al, 1993). This suggestion was supported by results showing that clusterin was unable to independently facilitate the reactivation of heat- inactivated ADH and catalase after the removal of stress (Figure 3.3.9). However, in the presence of a chaperone with refolding capability, Hsc70, and ATP, clusterin- stabilised ADH and catalase were partially refolded (Figure 3.3.9).

Results presented in 4.3.6 demonstrate that clusterin was effective in suppressing the slow precipitation of y-crystallin and lysozyme, but was unable to prevent these same target proteins from rapid precipitation (Figure 4.3.8). Real-time

H NMR spectroscopic analysis of the interaction between clusterin and a-lactalbumin

revealed that clusterin specifically interacts with the disordered molten globule (I2) state of this protein on its off-folding pathway (Figure 4.3.9). In addition, results presented in Poon et al, 2001, show that clusterin does not bind to (i) native a- lactalbumin, (ii) its stable, monomeric, partly structured intermediates, or (iii) the fully unfolded form of a-lactalbumin but instead, binds only to the reduced form of a- lactalbumin which is 12-like. Taken together, these results indicate that the proteins most likely to interact with clusterin are those in a disordered molten globule state and slowly aggregating on the off-folding pathway.

A model for the involvement of clusterin during protein unfolding is shown in

Figure 6.1.1. Looking at this figure, it can be seen that this scheme is analogous to that proposed for the sHsps. According to this model, stressed proteins undergo unfolding

163 from the native conformation to the fully unfolded state via a series of intermediate

"molten globule" states which, depending on the degree of structure, can be classified

as being "ordered" (Ii) or "disordered" (I2). In the molten globule state, partially unfolded proteins expose significant hydrophobicity which causes them to become highly unstable and prone to aggregation and precipitation via the irreversible off- folding pathway. Like sHsps, clusterin prevents precipitation of these forms of proteins by binding them to form stable high molecular weight complexes. The non- native proteins are stabilised in a manner which allows for their subsequent refolding and reactivation by other classes of chaperones with refolding capabilities.

Protein-Unfolding

± I. ± I2 <— • U •

ADP + Pi -*—^ Protein Off-foldim Aggregation Pathway ATP Hsc70 Stable protein Precipitation complex Clusterin / sHsps

FIGURE 6.1.1: A schematic representation showing the proposed mechanism of chaperone action of clusterin during protein unfolding. Stressed proteins initially undergo protein unfolding via the classical protein-unfolding pathway (blue). Native (N) proteins begin by losing their structural conformation and adopting intermediary (molten-globule; MG) conformations. These unstable

intermediary states can be classified as being ordered (I,) or disordered (I2) depending on how much of the overall structure has been disrupted or unfolded. Sometimes the stressed proteins may become completely unfolded (U). Often however, before a protein has the chance to fully unfold, protein aggregation with adjacent polypeptides occurs which usually leads to the eventual irreversible precipitation of the proteins involved. To prevent this, chaperones such as clusterin (and sHsps) recognise and bind partly unfolded 'substrate' proteins to form stable protein complexes. Once bound, these substrate proteins are held in a state competent for subsequent refolding by another chaperone protein (e.g. Hsc70).

A number of reports characterising the interaction of sHsps with non-native proteins have noted that exposure of sHsps to elevated temperature induces significant

164 structural changes, resulting in (i) dissociation of sHsp oligomers, (ii) increased surface hydrophobicity, and (iii) enhanced binding to substrate proteins (Shearstone and Baneyx, 1999; Haslbeck et al, 1999). In contrast, elevated temperature does not result in significant changes to either the oligomeric state or chaperone activity of clusterin (Figures 4.3.2 & 4.3.3). However, mildly acidic conditions induce in clusterin a structural transformation and enhanced chaperone action; these changes are strikingly similar to those induced in sHsps by elevated temperature. At low pH, dissociation of clusterin oligomers was accompanied by increased exposure of hydrophobic regions (Poon et al, 2001), resulting in an enhanced ability of clusterin to inhibit the stress-induced precipitation of the substrate proteins tested (Figures 4.3.4 -

4.3.6). In a physiological context, this effect may be of significant importance. At sites of inflammation, cardiac ischaemia, infarcted brain, and in the brains of

Alzheimer's sufferers (McGeer and McGeer, 1997), a phenomenon known as acidosis occurs, where the local pH can fall below 6. Under these mildly acidic conditions, the enhancement of clusterin's chaperone action could help reduce the rate of progression or severity of pathology by inhibiting the aggregation and deposition of inflammatory and/or toxic insoluble protein deposits. If so, clusterin may have applications in medicine and disease therapy.

There is extensive evidence of a correlation between clusterin expression and different types of diseases (e.g. Alzheimer's disease), many of which arise as a result of protein misfolding and/or aggregation (see section 1.2.3). Although it is not clear what the exact function of clusterin or sHsps is in these pathological conditions, the many physiologically relevant similarities between clusterin and the sHsps (e.g. structural features, distribution, and relative abundance; refer to chapter 2) suggest that

165 its function in these diseases may be very similar to that of the sHsps. The demonstrated sHsp-like chaperone action of clusterin raises the possibility that clusterin may function in vivo as a molecular chaperone. This suggestion is supported by the ability of clusterin to inhibit the stress-induced precipitation of proteins in serum (Figure 3.3.2).

Clusterin expression has been correlated with cell survival as exemplified by the cyto-protective effects of the overexpression of clusterin by TNFa-treated cells

(Humphreys et al, 1997). In addition, it has been reported that purified clusterin added to medium containing cultured cells protects the cells from TNFa (Sintich et al,

1999) and from oxidative stress (Schwochau et al, 1998). Furthermore, when exogenous clusterin was added at 100 ug/ml (a physiologically relevant concentration) to the medium of human cell lines undergoing heat-stress (42°C for 44 h), it specifically increased cell survival by approximately 10% (S. Poon; unpublished results). Together, these results provide evidence that clusterin is capable of exerting cytoprotection by conferring thermotolerance to cells. In addition, these results suggest that clusterin may be able to confer protection when acting either intra- or extracellularly. However, as yet, the biological site(s) at which clusterin exerts its chaperone action have not been identified. Nevertheless, if clusterin exerts an intracellular chaperone action under stress conditions, it would most certainly be able to act in concert with ATP-dependent chaperones such as Hsc70 to stabilise and facilitate correct refolding of stressed proteins. This would provide a mechanism of cytoprotection similar to that proposed for the sHsps (Lee et al, 1997). If clusterin operates as an extracellular chaperone, a potential mechanism of action could involve the stabilisation and prevention of aggregation of stressed proteins (such as receptors) on the cell surface. This action could promote cell survival by minimising stress-

166 induced aberrant signalling. The ability of clusterin to cooperate in vitro with Hsc70 to facilitate refolding of stressed proteins suggests that in vivo, clusterin-stabilised stressed surface proteins might be refolded to regain normal function by the action of as yet unidentified ATP-dependent chaperones. In this context, it is noteworthy that oxidative and osmotic stresses are known to induce the release of ATP from cells to the extracellular space. It is therefore possible that stresses might induce sufficient release and accumulation of ATP in the extracellular space close to the stressed cell membranes to facilitate ATP-dependent refolding of stressed proteins.

At present, the regions of clusterin responsible for its chaperone action have not been identified. In addition, no X-ray crystallography data are available to provide clues about the three-dimensional structure of the protein. However, sequence analysis of clusterin has revealed several regions of structure that might be important in its function (de Silva et al, 1990b). Of particular interest to the study of the structure- function relationship of clusterin are five predicted regions of amphipathic a-helices and two predicted regions of coiled-coil structures (Figure 5.1.1). Amphipathic a- helices are thought to be important in mediating interactions with hydrophobic molecules. As clusterin has been shown to bind exposed regions of hydrophobicity on partially unfolded proteins (Hochgrebe et al, 2000; Poon et al, 2002), it is likely that the amphipathic a-helices are involved in the chaperone action of clusterin.

A recent study has presented results to suggest that regions important for the chaperone action of clusterin may be located towards the N- and C-terminus of the a- and p-chain of the protein, respectively (Lakins et al, 2002). Interestingly, these regions were predicted to contain three of the five amphipathic a-helices found in clusterin, as well as the two coiled-coil a-helices (Figure 5.1.1). Sequence studies

167 have revealed that parts of these regions, involving the predicted a-helices, are probably natively disordered and molten globule-like (Bailey et al, 2001). In addition, the binding of the hydrophobic probe l-anilino-8-naphthalenesulfonate (ANS) to these regions indicates their hydrophobic nature. Hydrophobic interactions are known to be important in the binding of chaperones to partially unfolded, disordered molten globule substrates. Therefore, based on the evidence provided above, it is reasonable to suggest that the regions of clusterin responsible for its chaperone function lie towards the N- and C-termini of the a and p chains (i.e. the predicted regions of disorder and hydrophobicity), probably within the predicted a-helical regions.

To test the validity of this statement, a series of recombinant human clusterin mutants were produced using site-directed mutagenesis and Gateway™ cloning technology, and expressed in Spodoptera frugiperda 1PLB-Sf9 (Sf9) insect cells (refer to Chapter 5). Specific mutations were introduced into the cDNA of clusterin to encode mutants in which (i) an amino acid within each of the amphipathic a-helices is changed to a proline, a procedure which should disrupt the secondary structure of the a-helices, and (ii) parts of the N- and C-termini of the a and P chains are truncated.

As results in Figure 5.3.6 show, the expressed products, including wild type recombinant clusterin, were approximately 15 kDa smaller than their expected sizes, indicating that they were probably incompletely glycosylated (section 5.3.5). In addition, contrary to numerous reports, results also indicate that transiently transfected

Sf9 insect cells were unable to produce high levels of recombinant clusterin.

However, in this laboratory, transfected CHO cells have recently been shown to express properly processed clusterin at levels much higher than that obtained with transiently transfected Sf9 cells. Thus, expression of wild-type and mutant clusterin

168 for use in future studies of clusterin's structure-function relationship will be conducted using an appropriate mammalian expression system.

Clearly, much remains to be done to clarify the cellular site(s) at which the chaperone action of clusterin is physiologically active and to better define the molecular mechanism(s) by which clusterin protects proteins and cells from stress.

Future studies will also have to examine the role of clusterin in the numerous diseases with which it is associated and from this, determine whether clusterin may have therapeutic applications. Furthermore, it will be necessary to determine the 3- dimensional structure (either by X-ray crystallography or NMR spectroscopy) of clusterin to better understand what structural features of the protein allow it to function as a molecular chaperone. Hopefully, expression and further studies of the various clusterin mutants will also provide valuable information to facilitate identification of the region(s) of the protein responsible for the chaperone action.

169 References

170 Ackerman, M.J., and Clapham, D.E. (1997) "Ion channels- Basic science and clinical disease" New Eng. J. Med: 336, 1575-1586.

Agawal, N., Jomary, C, Jones, S.E., O'Rourke, K., Chaitin, M., Wordinger, R.J., and Murphy, B.F. (1996) "Immunocytochemical co-localization of clusterin in apoptotic photoreceptor cells in retinal degeneration slow rds mutant mouse retinas" Biochem. Biophys. Res. Comm: 225, 84-91.

Ahuja, H.S., Tenniswood, M., Lockshin,R., and Zakeri, Z.F. (1994) "Expression of clusterin in cell differentiation and cell death" Biochem. Cell Biol: 72, 523-529.

Akesson, P., Sjoholm, A.G., and Bjorck, L. (1996) "Protein SIC, a novel extracellular protein of Streptococcus pyogenes interfering with complement function" /. Biol. Chem: 271, 1081-1088.

Amersham pharmacia biotech (2000) "The recombinant protein handbook: Protein purification and simple purification" Published by Amersham pharmacia biotech Inc. NJ, USA.

Anfinsen, C.B. (1972) "Studies on the principles that govern the folding of protein chains" Science: 181, 223-230.

Anfinsen, C.B., and Haber, E., (1961) "Studies on the reduction and re-formation of protein disufide bonds" J. Biol. Chem: 236, 1361-1363.

Aronow, B.J., Lund, S.D., Brown, T.L., and Harmony, J.A.K. (1993) "Apolipoprotein J expression at fluid-tissue interfaces: Potential role in barrier cytoprotection" Proc. Natl

Acad. Sci: 90, 725-729.

171 Bailey, R.W., Dunker, K.A., Brown, C.J., Garner, E.C., and Griswold, M.D. (2001) "Clusterin, a binding protein with molten globule-like region" Biochemistry: 40, 11828-11840.

Balbach, J., Forge, V., Lau, W.S., van Nuland, N.A.J., Brew, K., and Dobson, CM. (1996) "Protein folding monitored at individual residues during two-dimensional NMR experiment. Science: 274, 1161.

Balbach, J.J., Ishii, Y., Antzutkin, N.O., Leapman|, R.D., Rizzo, N.W., Dyda, F., Reed,

J., and Tycko, R. "Amyloid Fibril Formation by APi6.22, a Seven-Residue Fragment of the Alzheimer'sP-Amyloid Peptide, and Structural Characterization by Solid State NMR" Biochemistry: In press.

Bettuzzi, S., Hiipakka, R.A., Gilna, P., and Liao, S. (1989) "Identification of an -repressed mRNA in rat ventral prostate as coding for sulphated glycoprotein 2 by cDNA cloning and sequence analysis" Biochem. J: 257, 293-296.

Bhattacharyya, J., and Das, K.P. (1999) "Molecular chaperone-like properties of an

unfolded protein, as- Casein" /. Biol Chem: 274, 15505-15509.

Blaschuk, O., Burdzy, K., and Fritz, LB. (1983) "Purification and characterization of a cell-aggregating factor (Clusterin), the major glycoprotein in ram rete testis fluid". J. Biol Chem: 12,7714-7720.

Boggs, L.N., Fuson, K.S., Baez, M., Churgay, L., McClure, D., Becker, G., and May, P.C. (1996) "Clusterin (ApoJ) Protects against In Vitro Amyloid-(1-40) Neurotoxicity"/ Neurochem: 67, 1324-1327.

Bora, N.S., Gobleman, C.L., Atkinson, J.P., Pepose, J.S., and Kaplan, H.J. (1993) "Differential expression of the complement regulatory proteins in the human eye"

Invest. Ophthalm. Vis. Sci: 34, 3579-3584.

172 Branden, C, and Tooze, J., (1999) "Introduction to protein structure, 2nd Ed." Garland Publishing, Inc. NY, USA.

Burkey, B.F., de Silva, H.V., and Harmony, J.A.K. (1991) "Intracellular processing of apolipoprotein J precursor to the mature heterodimer" /. Lip. Res: 32, 1039-1048. Buttyan, R., Olsson, C.A., Pintar, J., Chang, C, Bandyk, M., Ng, P-Y., and Sawczuk, I.S. (1989) "Induction of the TRPM-2 gene in cells undergoing programmed death" Molec. Cell. Biol: 9, 3473-3481.

Calero, M., Rostagno, A., Matsubara, E., Zlokovic, B., Frangione, B., and Ghiso, J. (2000) "Apolipoprotein J (clusterin) and Alzheimer's disease" Microscopy Res. Tech: 50,305-315.

Calvo, E.L., Mallo, G.V., Fiedler, F., Malka, D., Vaccaro, M.I., Keim, V., Morisset, J., Dagorn, J-C, and Iovanna, J.L. (1998) "Clusterin overexpression in rat pancreas during the acute phase of pancreatitis and pancreatic development" Eur. J. Biochem: 254, 282- 289.

Carver, J.A., Guerreiro, N., Nicholls, K.A., and Truscott, R.J.W. (1995), "On the interaction of a-crystallin with unfolded protein." Biochim. Biophys. Acta. 1252, 251-

260.

Carver, J.A., Lindner, R.A., Lyon, C, Canet, D., Hernandez, H., Dobson, CM., and Redfield, C (2002) "The interaction of the molecular chaperone a-crystallin with unfolded a-lactalbumin: a structure and kinetic spectroscopic study" J. Mol Biol: 318,

815-827.

Chang, Z., Primm, T.P., Jakana, J., Lee, I.H., Serysheva, I., Chiu, W., Gilbert, H.F., and Quiocho, F.A. (1996) "Mycobacterium tuberculosis 16-kDa antigen (Hspl6.3) functions as an oligomeric structure in vitro to suppress thermo aggregation. J. Biol.

Chem. 271,7218-7223. Choi, N-H., Mazda, T., and Tomita, M. (1989) "A serum protein SP40,40 modulates the formation of membrane attack complex of complement on erythrocytes" Molec. Immunol: 26, 835-840.

Choi-Miura, N.H., and Oda, T. (1996) "Relationship between multifunctional protein "Clusterin" and Alzheimer disease" Neurobiol of Aging: 17, 717-722. Choi-Miura, N.H., Takahashi, Y., Nakano, Y., Tobe, T., and Tomita, M. (1992), "Identification of the disulfide bonds in human plasma protein SP-40,40 (Apolipoprotein-J)" J. Biochem: 112, 557-561.

Clark, A.M., and Griswold, M.D. (1997) "Expression of clusterin/sulfated glycoprotein-2 under conditions of heat stress in rat Sertoli cells and a mouse Sertoli cell line" J. Androl: 18,257-263.

Collard, M.W., and Griswold, M.D. (1987) "Biosynthesis and molecular cloning of sulfated glycoprotein 2 secreted by rat Sertoli cells" Biochemistry: 26, 3297-3303.

Courgeon, A-M, Rollet, E., Becker, J., Maisonhaute, C, and Best-Belpomme, M. (1988), " Hydrogen peroxide (H2O2) induces actin and some heat-shock proteins in Drosophila cells." Eur. J. Biochem. 171, 163-170.

Creighton, T.E., (1997) "Protein folding coupled to disulphide bond formation" Biol Chem: 378,731.

Creighton, T.E., (1997) "Protein function; a practical approach" Oxford University Press, Oxford, UK.

Dandoy-Dron, F., Guillo, F., Benboudjema, L., Deslys, J-P., Lasmezas, C, Dormont, D., Tovey, M.G., and Dron, M. (1998) "Gene expression in Scrapie" J. Biol Chem:

273, 7691-7'697.

Danik, M., Chabot, J-G., Mercier, C, Benabid, A.L., Chauvin, C, Quirion, R., and Suh, M. (1991) "Human gliomas and epileptic foci express high levels of a mRNA

174 related to rat testicular sulfated glycoprotein 2, a purported marker of cell death" Proc. Natl Acad. Sci: 88, 8577-8581.

Data, S.A., and Rao, CM. (1999) "Differential temperature-dependent chaperone-like activity of aA- and aB-crystallin homoaggregates" J. Biol. Chem: 274, 34773-34778. de Silva, H.V., Stuart, W.D., Park, Y.B., Moa, S.J.T., Gil, CM., Wetterau, J.R., Busch, S.J., and Harmony, J.A.K. (1990a) "Purification and characterization of Apolipoprotein J". /. Biol. Chem: 265, 14292-14297. de Silva, H.V., Harmony, J.A.K, Stuart, W.D., Gil, CM., and Robbins, J., (1990b) "Apolipoprotein J: Structure and tissue distribution" Biochemistry: 29, 5380-5389. de Silva, H.V., Stuart, W.D., Duvic, C.R., Wetterau, J.R., Ray, M.J., Ferguson, D.G., Albers, H.W., Smith, W.R. and Harmony, J.A.K. (1990c) "A 70-kDa Apolipoprotein designated ApoJ is a marker for subclasses of human plasma high density lipoproteins" J Biol Chem: 265, 13240-13247.

Diemer, V., Hoyle, M., Baglioni, C, and Millis, A.J.T. (1992) "Expression of porcine complement cytolysis inhibitor mRNA in cultured aortic smooth muscle cells" J. Biol. Chem: 267, 5257-5264.

Dill, K.A., (1990) "Dominant forces in protein folding" Biochemistry: 29, 7133-7155. Dobson, CM. (1999) "Protein misfolding, evolution and disease" Trends Biochem.

Sci: 9, 329.

Dota, A., Nishida, K, Quantock, A.J., and Kinoshita, S. (1999) "Clusterin in human corneal endothelium and aqueous humor" Exp. Eye Res: 69, 705-708.

Dreusicke, D., and Schulz, G.E., (1988) "A switch between two conformations of adenylate kinase" J. Mol. Biol: 203, 1021-1028.

175 Duguid, J.R., Bohmont, C.W., Liu, N., and Tourtellotte, W.W. (1989) "Changes in brain gene expression shared by scrapie and Alzheimer disease" Proc. Natl. Acad. Sci: 86,7260-7264.

Dunker, A.K., Lawson, J.D., Brown, C.J., Williams, R.M., Romero, P., Oh, J.S., Oldfield, C.J., Campen, A.M., Ratliff, CM, Hipps, K.W., Ausio, J, Nissen, M.S., Reeves, R., Kang, C, Kissinger, C.R., Bailey, R.W., Griswald, M.D., Chiu, W., Garner, E.C, and Obradovic, Z, (2001) "Intrinsically disordered protein" J. Mol Graph. Modelling: 19, 26-59.

Ehrnsperger, M, Graber, S, Gaestel, M, and Buchner, J. (1997) "Binding of non- native protein to HSP25 during heat shock creates a reservoir of folding intermediates for reactivation" EMBO J: 16, 221-229.

Ellis, J, (1987) "Proteins as molecular chaperones." Nature: 328, 378-379.

Ellis, J., (1992) "Cytosolic chaperones confirmed." Nature: 358, 191-192.

Ellis, J, and Haiti, F-U. (1996) "Protein folding in the cell: competing models of function" FASEB J: 10, 20-26.

Farahbakhsh, Z.T., Huang, Q-L, Ding, L,L, Altenbach, C, Steinhoff, H.J, HorwitzJ, and Hubbell, W.L. (1995) "Interaction of a-crystallin with spin labelled peptides" Biochemistry: 34, 509-516.

Fersht, A.R. (1993) "Protein folding and stability- the pathway of folding barnase" FEBS Lett: 2,25, 5.

Fersht, A.R., Itzhaki, L.S, El Masry, N.F, Matthews, J.M, and Otzen, D.E. (1994) "Single versus parallel pathways of proteinfolding and fractional formation of structure in the transition state" Proc. Natl. Acad. Sci., USA: 91, 10426.

176 Fink, T.M, Zimmer, M, Tschopp, J, Etienne, J, Jenne, D.E, and Lichter, P. (1993) "Human clusterin (CLI) maps to 8p21 in proximity to the lipoprotein lipase (LPL) gene" Genomics: 16, 526-528.

Fischer-Colbrie, R, Zangerle, R, Frischenschlager, I, Weber, A, and Winkler, H. (1984) "Isolation and immunological characterization of a glycoprotein from adrenal chromaffin granules" J. Neurochem: 42, 1008-1016.

French, L.E, Wohlwend, A, Sappino, A.P, Tschopp, J, and Schifferli, J.A. (1994) " Human clusterin gene expression is confined to surviving cells during in vitro programmed cell death" J.Clin.Invest: 93, 877-884.

Fritz, LB, Burdzy, K, Setchell, B, and Blaschuk, O. (1983) "Ram rete testis fluid contains a protein (clusterin) which influences cell-cell interactions in vitro". Biology of reproduction: 28, 1173-1188.

Gamier. J, Gibrat. J, and Robson. B. (1996) "GOR secondary structure prediction method version IV" Methods in Enzymology: 266, 540-553.

Garret, R, and Grisham, CM. (1995), "Biochemistry: Textbook chaperones" Saunders Publishers, USA.

Gelissen, I.C, Hochgrebe, T, Wilson, M.R, Easterbrook-Smith, S.B, Jessup, W, Dean, R.T, and Brown, A.J. (1998) "Apolipoprotein J (clusterin) induces cholesterol export from macrophage-foam cells: a potential anti-atherogenic function?" Biochem.

J: 331, 231-237.

Gething, M.J, 1997, "Guidebook to molecular chaperones and protein- folding catalysis." Sambrook and Tooze Pub. NY, USA. Pp.269-272. Gething, M-J. (1997) "Guidebook to molecular chaperones and protein-folding catalysis" Sambrook and Tooze Publications, Oxford, UK.

Ghiso, J, Matsubara, E, Koudinov, A, Choi-Miura, N.H, Tomita, M, Wisniewski, T, and Frangione, B. (1993) "The cerebral-fluid soluble form of Alzheimer's amyloid beta

177 is complexed to SP-40, 40 (apolipoprotein J), an inhibitor of the complement membrane-attack complex" Biochem. J: 293, 27-30.

Goldberger, R.F, Epstein, CJ, and Anfinsen, CB, (1963) " Acceleration of reactivation of reduced bovine pancreatic ribonuclease by a microsomal system from rat liver" J. Biol Chem: 238, 628-635.

Goldner-Sauve, A, Szpirer, C, Szpirer, J, Levan, G, and Gasser, D.L. (1991) "Chromosome assignments of the genes for glucocorticoid receptor, myelin basic protein, leukocyte common antigen, and TRPM2 in the rat" Biochemical genetics: 29, 275-286.

Griko, Y.V, and Remeta, D.P. (1999) "Energetics of solvent and ligand-induced conformational changes in alpha-lactalbumin" Protein Sci: 8, 554-561.

Griswold, M.D, Roberts, K, and Bishop, P. (1986) "Purification and characterization of a Sulfated Glycoprotein secreted by Sertoli cells" Biochemistry: 25, 7265-7270.

Hara-Kuge, S, Ohkura, T, Ideo, H, Shimada, O, Atsumi, S, and Yamashita, K. (2002) "Involvement of VJP36 in intracellular transport and secretion of glycoproteins in polarized Madin-Darby canine kidney (MDCK) cells" J. Biol Chem: 277, 16332-

16339.

Hartl, F.U. (1996) "Molecular chaperones in cellular protein folding" Nature: 381,

571-580.

Hartmann, K, Rauch, J, Urban, J, Parczyk, K, Did, P, Pilarsky, C, Appel, D, Haase, W, Mann, K, Weller, A, and Koch-Brandt, C (1991) "Molecular cloning of gp 80, a glycoprotein complex secreted by kidney cells in Vitro and in Vivo" J. Biol.

Chem: 266,9924-9931.

Haslbeck, M, Walke, S, Stromer, T, Ehrnsperger, M, White, H.E, Chen, S, Saibil, H.R, and Buchner, J. (1999) "Hsp26: a temperature-regulated chaperone" EMBO J:

18,6744-6751. Hatters, D.M, Wilson, M.R, Easterbrook-Smith, S.B, and Howlett, G.J. (2002) "Suppression of apolipoprotein CU amyloid formation by the extracellular chaperone, clusterin" Eur. J. Biochem: 269, 1-6.

Hee Han, B, DeMattos R.B, Dugan, L.L, Kim-Han, J.S, Brenza, R.P, Fryer, J.D, Kierson, M, Cirrito, J, Quick, K, Harmony, J.A.K, Aronow, B.J, and Holtzman, D.M. (2001) "Clusterin contributes to caspase-3-independent brain injury following neonatal hypoxia-ischemia" Nature Medicine: 7, 338-343.

Heininger, K. (1999) "A unifying hypothesis of Alzheimer's Disease. U. Pathological processes" Hum. Psychopharmacol. Clin. Exp: 14, 525-581.

Hermo, L, Lustig, M, Lefrancois, S, Argraves, W.S, and Morales, C.R. (1999) "Expression and regulation of LRP-2/megalin in epithelial cells lining the efferent ducts and epididymis during postnatal development" Mol Reprod. Develop: 53, 282- 293.

Herzfeld, J, (1991) "Understanding hydrophobic behaviour" Science: 253, 88.

Hochgrebe, T, Pankhurst, G.J, Wilce, J, and Easterbrook-Smith, S.B. (2000) "pH- dependent changes in the in vivo ligand-binding properties and structure of human clusterin" Biochemistry: 39, 1411-1419.

Holers, V.M, Cole, J.L, Lubin, D.M, Seya, T, and Atkinson, J.P. (1985) "Human C3b- and C4b-regulatory proteins: a new multi-gene family" Immunol. Today: 6, 188- 192.

Horwitz, J. (1992), " a-crystallin can function as a molecular chaperone." Proc. Natl.Acad. Sci, USA. 89, 10449-10453.

Horwitz, J, Huang, Q.-L, Ding, L. and Bova, M. P. (1998) "Lens a-crystallin: chaperone-like properties" Methods Enzymol: 290, 365-383.

179 Humphreys, D, Hochgrebe, T.T., Easterbrook-Smith, S.B, Tenniswood, M.P, and Wilson, M.R. (1997) "Effects of clusterin overexpression on TNF-a- and TNF-p- mediated death of L929 cells" Biochemistry: 36, 15233-15243.

Humphreys, D.T, Carver, J.A, Easterbrook-Smith, S.B, and Wilson, M.R, (1999) "Clusterin has chaperone-like activity similar to that of the small heat shock proteins" /. Biol. Chem: 274, 6875-6881.

Jackson, S.E, and Fersht, A.R. (1991) "Folding of chymotrypsin inhibitor 2:1. Evidence for a two-state transition" Biochemistry: 30, 10428.

Jakob, U, and Buchner, J. (1994) "assisting spontaneity: the role of Hsp90 and small Hsps as molecular chaperones" TIBS: 19, 205-211.

Jakob, U, Gaestel, M, Engel, K, and Buchner, J. (1993) "Small heat shock proteins are molecular chaperones." J. Biol. Chem. 268: 1517-1520.

Jenne, D.E, and Tschopp, J. (1989c) "Molecular structure and functional characterization of a human complement cytolysis inhibitor found in blood and seminal plasma: Identity to sulfated glycoprotein 2, a constituent of rat testis fluid" Proc. Natl. Acad. Sci: 86,7123-7127.

Jenne, D.E, and Tschopp, J. (1992b) "Clusterin: the intriguing guises of a widely expressed glycoprotein." TIBS: 17, 154-159.

Jenne, D.E, Lowin, B, Peitsch, M.C, Bottcher, A, Schmitz, G, and Tschopp, J. (1991a) "Clusterin (complement lysis inhibitor) forms a high density lipoprotein complex with apolipoprotein A-I in human plasma." J. Biol. Chem: 266,11030-11036.

Jimenez, J.L, Guijarro, J.I, Orlova, E, Zurdo, J, Dobson, CM, Sunde, M. and Saibil, H. (1999) "Cryo-electron microscopy structure of an SH3 amyloid fibril and model of the molecular packing" EMBO J: 18, 815-821.

180 Jomary, C, Darrow, R.M, Wong, P, Organisciak, D.T, Neal, M.J, and Jones, S.E. (1999) "Lack of casual relationship between clusterin expression and photoreceptor apoptosis in light-induced retinal degeneration" J. Neurochem: 72, 1923-1929.

Jones, D, Russnak, R.H, Kay, R.J, and Candido, E.P.M (1986), "Structure, expression and evolution of a small heat shock gene locus in Caenorhabditis elegans that is flanked by repetitive elements." J. Biol. Chem., 261, 12006-12015.

Jones, S.E, Meerabux, J.M.A, Yeats, D.A, and Neal, M.J. (1992) "Analysis of differentially expressed genes in retinitis pignetosa retinas. Altered expression of clusterin mRNA" FEBS: 300, 279-282.

Jordan-Starck, T, Lund, S.D, Wiite, D.P, Aronow, B.J, Ley, CA, Stuart, W.D, Swertfeger, D.K, Clayton, L.R, Sells, S.F, Paigen, B, and Harmony, J.A.K. (1994) "Mouse apolipoprotein J: Characterization of a gene implicated in atherosclerosis." J. Lip. Res: 35, 194-210.

Jordan-Starck, T.C, Witte, D.P, Aronow, B.J, and Harmony, J.A.K. (1992) "Apolipoprotein J: a membrane policeman?" Curr. Opin. Lipid: 3, 75-85.

Kabakov, A.E, and Gabai, V.L. (1997) "Heat shock proteins and cytoprotection" Springer Publishers, Heidelberg, Germany.

Kapito, R.R, and Ron, D. (2000) "Conformational disease" Nature Cell Biology: 2,

207-209.

Kapron, J.T, Hilliard, G.M, Lakins, J.N, Tenniswood, M.P.R, West, K.A., Carr, S.A, and Crabb, J.W. (1997) "Identification and characterization of glycosylation sites in human serum clusterin" Pro. Sci: 6, 2120-2133.

Kelso, G.J, Stuart, W.D, Richter, R.J, Furlong, C.E., Jordan-Starck, T.C, and Harmony, J.A.K. (1994) "Apolipoprotein J is associated with paraoxonase in human plasma" Biochmeistry: 33, 832-839.

181 Kim, KK, Kim, R, and Kim, S.H. (1998), "Crystal structure of a small heat shock protein. Nature, 394, 595.

Kim, K-S, and Woodward, C, (1993) "Protein internal flexibility and global stability: effect of urea on hydrogen exchange rates of bovine pancreatic trypsin inhibitor" Biochemistry: 32, 9609-9613.

Kim, R, Kim, KK, Yokota, H, and Kim, S.H. (1998) "Small heat shock protein of Methanococcus jannaschii, a hyperthermophile." Proc. Natl. Acad. Sci. USA, 95, 9129- 9133.

Kirszbuam, L, Sharpe, J.A, Murphy, B, d'Apice, A.J.F, Classon, B, Hudson, P, and Walker, I.D. (1989) "Molecular cloning and characterization of the novel, human complement-associated protein, SP-40, 40: a link between the complement and reproductive systems" EMBO J: 8, 711-718.

Kleizen, B, Braakman, I, and de Jonge, H.R. (2000) "Regulated trafficking of the CFTR chloride channel" Eur. J. Cell Biol: 79, 544-556.

Klemenz, R, Frohli, E, Steiger, R.H, Schafer, R, and Aoyama, A, (1991) "aB- crystallin is a small heat shock protein." Proc. Natl. Acad. Sci. 88:3652-3656.

Kounnas, M.Z., Loukinova, E.B, Stefansson, S, Harmony, J.A.K, Brewer, B.H, Strickland, D.K, and Argraves, W.S. (1995) "Identification of glycoprotein 330 as an endocytic receptor for apolipoprotein J/clusterin" J. Biol. Chem: 270, 13070-13075.

Kuby, J, (1997), "Immunology" 3rd Ed, Chap. 14, W.H. Freeman and Company, NY,

USA.

Lakins, J.N, Poon, S, Easterbrook-Smith, S.B, Carver, J.A, Tenniswood, M.P.R, and Wilson, M.R. (2002) "Evidence that clusterin has discrete chaperone and ligand binding sites" Biochemistry: 41, 282-291.

182 Laskey, R.A, Honda, B.M, Mills, A.D, and Finch, J.T. (1978) "Nucleosomes are assembled by an acidic protein which binds histones and transfers them to DNA" Nature: 275, 416-420.

Law, G.L, and Griswold, M.D. (1994) "Activity and form of sulfated glycoproetin 2 (clusterin) from cultured Sertoli cells, testis, and epididymis of the rat" Biol Reprod: 50, 669-679.

Lee, G.J, Roseman, A.M., Saibil, H.R, and Vierling, E. (1997) "A small heat shock protein stably binds heat-denatured model substrates and can maintain a substrate in a folding-competent state" EMBO J: 16, 659-671.

Lee, J.P, Pokala, N, and Vierling, E. (1995) "Structure and in vitro molecular chaperone activity of cytosolic small heat shock proteins from pea." J. Biol. Chem. 270: 10432-10438.

Leger, J.G, Montpetit, M.L, Tenniswood, M.P, (1987) Characterization and cloning of androgen-repressed mRNAs from rat ventral prostate" Biochem. Biophys. Res. Comm: 147, 196-203.

Leroux, M.R, Melki, R, Gordon, B, Batelier, G, and Candido, E.P.M, (1997) "Structure-function studies on small heat shock protein oligomeric assembly and interaction with unfolded polypeptides." J. Biol. Chem. 272: 24646-24656.

Lindner, R. A, Kapur, A. and Carver, J. A. (1997) "The interaction of the molecular chaperone, a-crystallin, with molten globule states of a-lactalbumin" J. Biol Chem: 111, 11111-17719.

Lindner, R. A, Kapur, A, Mariani, M, Titmuss, S. J. and Carver, J. A. (1998) "Structural alterations of a-crystallin during its chaperone action" Eur. J. Biochem:

258, 170-183.

183 Lindner, R.A, Treweek, T.M, and Carver, J.A. (2001) "The molecular chaperone a- crystallin is in kinetic competition with aggregation to stabilize a monomeric molten- globule form of a-lactalbumin" Biochem. J: 354, 79-87.

Liszewski, K, Post, T.W, and Atkinson, J.P. (1991) "Membrane cofactor protein (MCP or CD46): Newest member of the regulators of complement activation gene cluster" Annu. Rev. Immunol: 9, 431-455.

Luckow, V.A, and Summers, M.D. (1988) "Trends in the development of baculovirus expression vectors" Biotechnology: 6, 47-55.

Mahon, M.G, Lindstedt, K.A, Hermann, M, Nimpf, J, and Schneider, W.J. (1999) "Multiple involvement of clusterin in chicken ovarian follicle development" J. Biol. Chem: 274, 4036-4044.

Matsubara, E, Frangione, B, and Ghiso, J. (1995) "Characterization of apolipoprotein J-Alzheimer's Ab interaction" J. Biol Chem: 270, 7563-7567.

Matthews, C.R. (1993) "Pathways of protein folding" Annu. Rev. Biochem: 62, 653- 683.

May, P.C, and Finch, CE. (1992) "Sulfated Glycoprotein 2: new relationships of this multifunctional protein to " TIBS: 15, 391-395.

May, P.C, Johnson, S.A, Poirier, J, Lampert-Etchells, M, and Finch, CE. (1989) "altered gene expression in Alzheimer's disease brain tissue" Can. J. Neurol. Sci: 16,

473-476.

May, P.C, Lampert-Etchells, M, Johnson, S.A, Poirier, J, Masters, J.N, and Finch, CE. (1990) "Dynamics of gene expression for a hippocampal glycoprotein elevated in Alzheimer's disease and in response to experimental lesions in rat" Neuron: 5, 831-

839.

184 McDonald, J.F, and Nelsustuen, G.L. (1997) "Potent inhibition of terminal compliment assembly by clusterin: characterisation of its impact on C9 polymerization" Biochemistry: 36, 7464-7473.

McGeer, E.G., and McGeer, P.L. (1997) "Innate inflammatory reaction of the brain in Alzheimer's disease" McGill J. Med: 3, 134-141.

McHattie, S, and Edington, N. (1999) "Clusterin prevents aggregation of neuropeptide 106-126 in vitro" Biochem. Biophys. Res. Commun: 259, 336-340.

Mehlen, P, Mehlen, A, Guillet, D, Preville, X, and Arrigo, A-P. (1995), " -a induces changes in the phosphorylation, cellular localization and oligomerization of human hsp27, a stress protein that confers cellular resistance to this cytokine." J. Cell. Biochem. 58, 248-259.

Mehlen, P, Preville, X, Chareyron, P, Briolay, J, Klemenz, R, and Arrigo, A.P. (1995) "Constitutive expression of human hsp27, Drosophila hsp27, or human aB- crystallin confers resistance to TNF- and oxidative stress- induced cytotoxicity in stably transfected murine L929 fibroblasts" J. Immunol: 154, 363-374.

Merck, KB, Groenen, P.J, Voorter, CE, de Haard-Hoekman, W.A, Horwitz, J, Bloemendal, H, and de Jong, W.W. (1993). "Structural and functional similarities of bovine a-crystallin and mouse small heat-shock protein. A family of chaperones." J.

Biol Chem. 268, 1046-1052.

Millis, A.J.T, Luciani, M, McCue, H.M, Rosenberg, M.E, and Moulson, C.L. (2001) "Clusterin regulates vascular smooth muscle cell nodule formation and migration" J.

Cell Physiol: 186, 210-219.

Morales, C.R., Igdoura, S.A, Wosu, U.A, Boman, J, and Argraves, W.S. (1996) "Low density lipoprotein receptor-related protein-2 expression in efferent duct and epididymal epithelial: evidence in rats for its in vivo role in endocytosis of apolipoprotein J/clusterin" Biol Reprod: 55, 676-683.

185 Moulson, CL, and Millis, A.J.T. (1999) "Clusterin (Apo J) regulates vascular smooth muscle differentiation in vitro" J. Cell. Physiol: 180, 355-364.

Muchowski, P.J, Wu, G.J, Liang, J.J, Adman, E.T, and Clark, J.I. (1999) "Site- directed mutations within the core 'a-crystallin' domain of the small heat-shock protein, human aB-crystallin, decrease molecular chaperone functions" J. Mol Biol: 289, 397.

Murphy, B.F, Kirszbaum, L, Walker, I.D., and d'Apice, A.J.F. (1988) "SP-40, 40, a newly identified normal human serum protein found in the SC5b-9 complex of complement and in the immune deposits in " J. Clin. Invest. 81, 1858-1864.

O'Bryan, M.K, Baker, H.W.G, Saunders, J.R, Kirszbaum, L, Walker, I.D, Hudson, P., Liu, D.Y, Glew, M.D, and d'Apice, A.J.F. (1990) "Human seminal clusterin (SP- 40,40), isolation and characterization" J. Clin Invest. 85, 1477-1486.

Oda, T, Pasinetti, G.M, Osterburg, H.H, Anderson, C, Johnson, S.A, and Finch, CE. (1994) "Purification and characterization of brain clusterin." Biochem. Biophys. Res. Commun: 204, 1131-1136.

Ohgushi, M, and Wada, A, (1983) "Molten-globule state: a compact form of globular proteins with mobile side chains" FEBS Letters: 164, 21-24.

Okazaki, A, Ikura, T, Nikaido, K, and Kuwajima, K. (1994) "The chaperone GroEL does not recognize apo-alpha-lactalbumin in the molten state" Nat. Struct. Biol: 1, 439- 446.

Pain, R.H, (2000) "Mechanisms of protein folding, 2nd Ed." Oxford University Press:

Oxford, UK.

186 Palmer, D.J, and Christie, D.L. (1990) "The primary structure of Glycoprotein m from bovine adrenal medullary chromaffin granules: Sequence similarity with human serum protein-40,40 and rat Sertoli cell glycoprotein" J. Biol. Chem: 265, 6617-6623.

Pankhurst, G.J, Bennett, CA, and Easterbrook-Smith, S.B. (1998) "Characterisation of the Heparin-binding properties of human clusterin" Biochemistry: 37, 4823-4830.

Parczyk, K, Pilarsky, C, Rachel, U, and Koch-Brandt, C (1994) "Gp80 (clusterin;TRPM-2) mRNA level is enhanced in human renal clear cell carcinomas" J. Cancer Res. Clin. Oncol: 120, 186-188.

Pauling, L, and Corey, R.B, (1951) "The pleated sheet, a new layer configuration of polypeptide chains" Proc. Natl. Acad. Sci., USA: 37, 251-256.

Pavlicek, Z, and Ettrich, R. (1999) "Chaperone-like activity of human : similarity with a-crystallin" Collect. Czech. Chem. Commun: 64,111-116. Pfeifer, T.A. (1998) "Expression of heterologous proteins in stable insect cell culture" Current Opinion in Biotechnology: 9, 518-521.

Pilarsky, C, Haase, W, and Koch-Brandt, C (1993) "Stable expression of gp80 (TRPM-2, clusterin), a secretory protein implicated in programmed cell death, in transfected BHK-21 cells" Biochim. Biophys. Acta: 1179, 306-310.

Plater, M.L, Goode, D, and Crabbe, J.C. (1996) "Effects of site-directed mutations on the chaperone-like activity of aB-crystallin" J. Biol Chem: 271, 28558-28566.

Poon, S, Easterbrook-Smith, S.B,Rybchyn, M.S., Carver, J.A, and Wilson, M.R. (2000) "Clusterin is an ATP-independent chaperone with very broad substrate specificity that stabilizes stressed proteins in a folding-competent state" Biochemistry:

39, 15953-15960.

187 Poon, S, Treweek, T.M, Wilson, M.R, Easterbrook-Smith, S.B, and Carver, J.A. (2002) "Clusterin is an extracellular chaperone that specifically interacts with slowly aggregating proteins on their off-folding pathway" FEBS Letters: 513, 259-266.

Poon, S, Rybchyn, M.S., Easterbrook-Smith, S.B, Carver, J.A, Pankhurst, G.J, and Wilson, M.R. (2002) "Mildly Acidic pH Activates the Extracellular Molecular Chaperone Clusterin" J. Biol. Chem: 277, 39532-39540.

Preville, X, Salvemini, F, Giraud, S, Chaufer, S, Paul, C, Stepien, G, Ursini, M.V, and Arrigo, A.P. (1999) "Mammalian small stress proteins protect against oxidative stress through their ability to increase glucose-6-phosphate dehydrogenase activity and by maintaining optimal cellular detoxifying machinery" Exp. Cell. Res: 247, 61-78.

Raman, B, and Rao, CM. (1997) "Chaperone-like activity and temperature-induced structural changes of a-crystallin" J. Biol. Chem: 272, 23559-23564.

Reddy, K.B, Karode, M.C, Harmony, J.A.K, and Howe, P.H. (1996) "Transforming Growth Factor (TGF)-Induced Nuclear Localization of Apolipoprotein J/Clusterin in Epithelial Cells" Biochemistry: 35, 6157-6163.

Rennie, P.S, Bruchovsky, N, Akakura, K, Goldenberg, L.S, Otal, N, Akakura, S, Wong, P, and Tenniswood, M. (1994) "Effect of tumour progression on the androgenic regulation of the androgen receptor, TRPM-2 and YPT1 genes in the shionogi carcinoma" J. Steroid Biochem. Molec. Biol: 50, 31-40.

Rosenberg, M.E, and Silkensen, J. (1995) "Clusterin: Physiologic and pathophysiologic considerations" Int. J. Biochem: Cell Biol 27, 633-645.

Rosenberg, M.E, Dvergsten, J, and Correa-Rotter, R. (1993) "Clusterin: An enigmatic protein recruited by diverse stimuli." J. Lab. Clin. Med: 111, 205-214.

Ruddon, R.W, and Bedows, E, (1997) "Assisted protein folding" J. Biol. Chem: 111,

3125-3128. Sali, A, Shakhnovich, E, and Kurplus, M. (1994) " How does a protein fold?" Nature: 369, 248.

Sambrook, J, Fritsch, E.F, Maniatis, T. (1989), "Molecular Cloning, A Laboratory Manual", 2nd Ed. Cold Spring Harbor Laboratory Press, USA.

Sanada, A, and Vo-Dinh, T. (2001) Biomedical implications of protein folding and misfolding" Biotechnol Appl. Biochem: 33, 7-16.

Sanchez, Y, and Lindguist, S.L (1990) "Hspl04 required for induced thermotolerance" Science: 248, 1112-1115.

Saunders, J.R, Aminian, A, Katrina, L.M, O'Farrell, A, Adam, W.R, and Murphy, B.F. (1994) "Clusterin depletion enhances immune glomerular injury in the isolated perfused kidney" Kidney International: 45, 817-827.

Schroder, H, Langer, T, Hartl, F-U, and Bukau, B. (1993) "DnaK, DnaJ and GrpE form a cellular chaperone machinery capable of repairing heat-induced protein damage" EMBO J: 12, 4137-4144.

Schulz, G.E, (1991) "Protein Conformation" John Wiley & Sons Publishing Ltd. West Sussex, England, UK (incorporates the works presented at the 161st Ciba Foundation Symposium, 22-24 January 1991).

Schwochau, G.B, Nath, K.A, and Rosenberg, M.E. (1998) "Clusterin protects against oxidative stress in vitro through aggregative and non-aggregative properties" Kidney

International: 53, 1647-1653.

Sensibar, J.A, Sutkowski, D.M, Raffo, A, Buttyan R, Griswold, M.D, Sylvester, S.R, Kozlowski, J.M, and Lee, C (1995) "Prevention of cell death induced by tumor necrosis factor a in LNCaP cells by overexpression of sulfated glycoprotein-2

(clusterin)" Cancer Res: 55, 2431-2437. Shearstone, J.R, and Baneyx, F. (1999) "Biochemical characterization of the small heat shock protein IbpB from Escherichia coli" J. Biol. Chem: 174, 9937-9945.

Shirley, B.A, Stanssens, P, Hahn, U, and Pace, CN, (1992) "Contribution of hydrogen bonding to the conformational stability of ribonuclease Tl" Biochemistry: 31, 725-732.

Silkensen, J.R, Skubitz, A.P.N, Skubitz, KM, and Rosenberg, M.E. (1999) "Identification of clusterin sequences mediating renal tubular cell interactions" J. Pep. Res: 54, 449-457.

Simons, J.F, Ferro-Novick, S, Rose, M.D, and Helenius, A. (1995) "Bip/Kar2p serves as a molecular chaperone during carboxypeptidase Y folding in yeast" J. Cell Biol: 130, 41-49.

Sintich, S.M, Steinberg, J, Kozlowski, J.M, Lee, C, Pruden, S, Sayeed, S, and Sensibar, J.A. (1999) "Cytotoxic sensitivity to tumor necrosis factor-a in PC3 and LNCaP cells is regulated by extracellular levels of SGP-2 (Clusterin)" The Prostate: 39, 87-93.

Sisodia, S.S, and Price, D.L. (1995) "Role of the p-amyloid protein in Alzheimer's disease" FASEB J: 9, 366-370.

Slepoy, A, Singh, R.R.P, Pazmandi, F, Kulkarni, R.N, and Cox, D.L. (2001) "On the statistical mechanisms of prion diseases" Article found at http://arvix.org/ps- cache/cond-mat/pdf/0102/0102478.pdf

Smith, P.K, Krohn, R.I, Hermanson, G.T, Mallia, A.K, Gartner, F.H, Provenzano, M.D, Fujimoto, E.K, Goeke, N.M, Olson, B.J, Klenk, D.C. (1985) "Measurement of protein using bicinchoninic acid" Anal. Biochem: 150, 76-85.

190 Stringham, E.G., and Candido, E.P.M. (1994), "Transgenic hspl6-lacZ strains of Caenorhabditis elegans as biological monitors of environmental stress." Environ. Toxicol. Chem. 13, 1211-1220.

Stryer, L. (1995) "Biochemistry" W.H. Freeman and Co., NY, USA.

Tenniswood, M, Wang, Z, Lakins, J, Morrissey, C, O'Sullivan, J, and Tang, H. (1998) "Clusterin in the male reproductive tract" J. Androl: 19, 508-514.

Thomas, P.J, Qu, B.H, and Pederson, P.L. (1995) "Defective protein folding as a basis of human disease" Trends Biochem. Sci: 20, 456.

Thomasson, W.A. (2001) "Unravelling the mystery of protein folding" Article found at http://www.faseb.org/opar/protfold/protfold.pdf

Treweek, T.M, Lindner, R.A, Mariani, M, and Carver, J.A. (2000), "The small heat- shock chaperone protein, alpha-crystallin, does not recognise stable molten globule states of cytosolic proteins." Biochim. Biophys. Acta. 1481, 175-188.

Tschopp, J, Chonn, A, Hertig, S, and French, L.E. (1993) "Clusterin, the human apolipoprotein and complement inhibitor, binds to complement C7, C8p, and the B- domain of C9" J. Immunology: 151, 2159-2165.

Tsuruta, J.K, Wong, K, Fritz, LB, and Griswold, M.D. (1990) " Structural analysis of sulfated glycoprotein 2" Biochem. J: 268, 571-578.

Urban, J, Parczyk, K, Leutz, A, Kayne, M, and Kondor-Koch, C. (1987) "Constitutive apical secretion of an 80-kD sulfated glycoprotein complex in the polarized epithelial Madin-Darby canine kidney cell line" J. Cell Biol: 105, 2735-2743.

Vakeva, A, Laurila, P, and Meri, S. (1993) "Co-deposition of clusterin with the complement membrane attack complex in " Immunology: 80,

177-182.

191 Van den Berg, B, Ellis, J.R, and Dobson, CM. (1999) "Effects of macromolecular crowding on protein folding and aggregation" EMBO J: 18, 6927-6933. van Holde, M, (1990) "Biochemistry" The Benjamin/Cummings Publishing Company: CA, USA.

Vialard, J.E, Arif, B.M, and Richardson, CD. (1995) "Methods in Molecular Biology; Baculovirus Expression Protocols" Edited by Richardson, CD. Humana Press Inc. NJ, USA

Viard, I, Wehrli, P, Jornot, L, Bullani, R, Vechietti, J-L, Schiferli, J.A, Tschopp, J, and French, L.E. (1999) "Clusterin gene expression mediates resistance to apoptotic cell death induced by heat shock and oxidative stress" J. Invest. Derma: 111, 290-296.

Waters, E.R, Lee, G.J, and Vierling, E. (1996) "Evolution, structure and function of the small heat shock proteins in plants" J. Exp. Bot: 47, 325-338.

Welch, W.J. (1985) "Phorbol ester, calcium ionophore, or serum added to quiescent rat embryo fibroblast cells all results in the elevated phosphorylation of two 28,000-dalton mammalian stress proteins." J. Biol. Chem. 260: 3058-3062.

Weston,T. (1985) "Atlas of anatomy" Marshall Cavendish Ltd., London, G.B. Wilson, M.R, and Easterbrook-Smith, S.B. (1992) "Clusterin binds by a multivalent mechanism to the Fc and Fab regions of IgG" Biochimica et Biophysica Acta: 1159,

319-326.

Wilson, M.R, and Easterbrook-smith, S.B. (1996) "Clusterin: A puzzling protein"

Today's Life Science. 26-31.

Wilson, M.R, and Easterbrook-Smith, S.B. (2000) "Clusterin is a secreted mammalian chaperone" TIBS: 25, 95-98.

Wong, P, Pineault, J, Lakins, J, Taillefer, D, Leger, J, Wang, C, and Tenninwood, M. (1993) "Genomic organization and expression of the rat TRPM-2 (Clusterin) gene, a gene implicated in apoptosis" J. Biol. Chem: 268, 5021-5031.

192 Zlokovic, B.V, Martel, C.L., Mackic, J.B, Matsubara, E, Wisniewski, T, McComb, G, Frangione, B, and hiso, J. (1994) "Brain uptake of circulating J and E complexed to Alzheimer's amyloid b." Biochem. Biophys. Res. Commun: 205, 1431-

1437.

Zwain, I, and Amato, P. (2000) "Clusterin protects granulosa cells from apoptotic cell death during follicular atresia" Exp. Cell Res: 257, 101-110. Appendix

194 The following is a collection of DNA sequences that were obtained by bi­ directional sequencing of wild type and mutated clusterin cDNA in the expression vector, pXTNsect-Dest38. Start codon is highlighted in green; stop codon is highlighted in red; the signal peptide is shown as green text; cDNAs corresponding to the a- and 0- chains are shown as blue and red text, respectively; and (where appropriate) sites where proline substitution has occurred is shown as underlined purple text. Letters other than A,C,G, or T, represent bases that could not be definitively identified by DNA sequencing.

pXINsect-Dest38clus WT

AAGTTTKTACAAAAAAGCAGGCTCGGCCACC^BATGAAGACTCTGCTGCTGTTTGTGG GGCTGCTGCTGACCTGGGAGAGTGGGCAGGTCCTGGGGGACCAAACGGTCTCAGACAAT GAGCTCCAGGAAATGTCCAATCAGGGAAGTAAGTACGTCAATAAGGAAATTCAAAATGC TGTCAACGGGGTGAAACAGATAAAGACTCTCATAGAAAANACAAACGAAGAGCGCAAGA CACTGCTCAGCAACCTAGAAGAAGCCAAGAAGAAGAAAGAGGATGCCCTAAATGAGACC AGGGAATCAGAGACAAAGCTGAAGGAGCTCCCAGGAGTGTGCAATGAGACCATGATGGC CCTCTGGGAAGAGTGTAAGCCCTGCCTGAAACAGACCTGCATGAAGTTCTACGCACGCG TCTGCAGAAGTGGCTCAGGCCTGGTTGGCCGCCAGCTTGAGGAGTTCCTGAACCAGAGC TCGCCCTTCTACTTCTGGATGARTGGTGACCGCATCGACTCCCTGCTGGAGAACGACCG GCAGCAGACGCACATGCTGGATGTCATGCAGGACCACTTCAGCCGCGCGTCCAGCATCA TAGACGAGCTCTTCCAGGACAGGTTCTTCACCCGGGAGCCCCAGGATACCTACCACTAC CTGCCCTTCAGCCTGCCCCACCGGAGGCCTCACTTCTWCTTTTMCCAAGTYCCGATCGT CCGAGCTTGATGCCTTCTCTCCGTACGAGCCCCTGAACTTCCACGCATGTTCCAGCCCT TCCTTRAACTGATACACGAGGCTCAGCAGGCCATGGACATCCACTTCCACAGCCCGGCC TTCCAAGCACCCGCCAACAGAATTCATACGAGAAGGCGACGATGACCGGACTGTGTGCC GGGAGGCTCAGGCCACTCACGGCTGCTGCCTGTTCCACCAACAACCCCTRCCAGGCTAA GCTGCGGCGGGAGCTCGACGAATCCCTCCAGGTCGCTGAGAGGTTGACCAGGAAATACA ACGAGCTGCTAAAGTCCTACCAGTGGAAGATGCTCAACACCTCCTCCTTGCTGGAGCAG CTGAACGAGCAGTTTAACTGGGTGTCCCGGCTGGCAAACCTCACGCAAGGCGAAGACCA GTACTATCTGCGGGTCACCACGGTGGCTTCCCACACTTCTGACTCGGACGTTCCTTCCG GTGTCACTGAGGTGGTCGTGAAGCTCTTTGACTCTGATCCCATCACTGTGACGGTCCCT GTAGAAGTCTCCAGGAAGAACCCTAAATTTATGGAGACCGTGGCGGAGAAAGCGCTGCA GGAATACCGCAAAAAGCACCGGGAGGAG| (GACCCAGCTMTCTTGTACAAAGTTGG pXINsect-Dest38clusTl 7

AACTTTGTACAAAAAAGCAGGCTCGGCCACCATCACTCASTMCGTCAATAAGGAAATTC AAAATGCTGTCAACGGGGTGACAACATAAAGACTCTTATAGAAAAAACAAACGAAGAGC SCAAGAYACTGCTCAGCAACMTAGAAGAAGCTAAGAAGAAGAAAGAGGATGTTTTAAAT GAGACCAGGGAATCAGAGACAAAGCTGAAGGAGCTCCCAGGAGTTGCAATGAGACCATG ATGGCCCTCTGGGAAGAGTGTAAGCCCTGCCTGAAACAGACCTGCATGAAGTTCTACGC ACGCGTCTGCAGAAGTGGCTCAGGCCTCCTTGGCCGCCAGCTTGAGGAGTTCCTGAACA GAGCTCGCCCTTGTACTTCTGGATGAATCGTSACCGCATCGACTCCCTCCTGGASAACG ACCGGCAGCASACGCACTATGTTSGATGTCATGCAGGAHTWCTTHAGCWGCGCGTNCAG CAYCATMGADGAKCTCTYCCAGGACASGYTCTTCACACGGGASCCCCAGGATACCTACC ACTACCTGCHCTTTAGCWTCCCCWATCGGAGGCCTCACTTCTTCTYTTCAAGTCCCGCA TCGTCCGCAGCTTGATGCCCTTCTCTCCGTACGAGCCCCTGAACTTCCACACCATGTTC CAGCCCTTCCTTGATATGATACACGCCGCCAACAGAAWTCATACGAGAAGGCGACGATG ACCGGACTGTGTGCCGGGAGATCCGCCACAACTCCACGGGCTGCTGCGGATGAAGGACC AGTGTGACAAGTGCCGGGAGATCTTGTCTGTGGACTGTTCCACCAACAACCCCTCCCAG GCTAACCTGCGGCGGGAGCTCGACGAATCCCTCCAGGTCGCTGAGAGGTTGACCAGAAA ATACAACGAGCTGCWAAAGTCCTACCAGTGGAAGATGCTCAACACCTCCTCCTTGCTGG AGCAGCTGAACGAACAGTTTAACTGGGTGTCCCAGCTAGCAAACCTCACGCAAAGCGAA GACCAGTACTATCTGCGGGTCACCACGGTGGCTTCCCACACTTCTGACTCGGACCTTCC TTCCGGTGTCACTGAGGTGTCGTGAAGCTCTTTGACTCTGATCCCATCACTGTGACGGT CCCTGTAGAAGTCTCCAGGAAGAACCCTAAATTTATGGAGACCGTGGCGGAGAAAGCCC TACAGGAATACCGCAAAAAGCACCGGGAGGAG| |GACCCAACTGTSTTGTAGAYYGTT GG

pXINsect-Dest38clusT78

AACTTTGTACAAAAAAGCAGGCTCGGCCACC^HGGAGTGTGCAATGAGACCAGGATGG CCCTCTGGGAAGAVTGTAAGCCCTGCCTGAAACAGACCTGCATGAAGTTCTACGCACGC GTCTGCAGAAGTGGCTCAVGCCTGGTTGGCCGCCAGCTTGAGGAGTTCCTGAACCAGAG CTCGCCCTTCTACTTCTGGATGAATGGTGACCGCATCGACTGGGTGCTGGAGAACGACC GGCAGCAGACGCACATGCTGGATGTCATGCAGGACCACTTCAVCCGCGCGTCCAGCATC ATAGACGAGCTCTTCCAGGACAGGTTCTTCACCCGGGAGCCCCAGGATACCTACCACTA CCTGCCCTTCAGCCTGCCCCACCGGAGGCCTCACTTCTTCTTTCCCAAGTCCCGCATCG TCCGCAGGYGGYGGCCCYGSTSGCCGGYCGGGCCCCMGYYSGYCCYCGCCATGTTCCAG CCCTTCTTAAGATGATACACGAGGCTCAGCAGGCCATGGACATCCACTTCCACAGCCCG GCCTTCCAGCACCCGCCAACAGAATTCATACGYGAAGGCGASGAMGACCGGACTGTGTG CCGGGAGATCCGCTAVAACTCMACGGGCTGCTGCGGWTMAAGGACCAGTGTMARAAGTG CCGGGWGMTTTTGTVTGTGGAMTGTMCCYCGMACAACGCSGGCCAGGRTAAGRTGCGGC GGGAGCTCGWVGAATVCCTCCAGGTMGMTGAGAGGTTGACCYGMMAATACHACGAGCTG CTAHAGTCCTACCAGTGGHAGATGCTCAACACCTCCTCCTTGCTGGAGCAGCTGMACGA GVAGTTTAACTGGGTGTCCCGGCTGGCAAACMTVASGCAAGGCGAAGACCAGTACTATC TGCGGGTCACSACGGTGGCTTCCCACACTTCTGAVTCGGACGTTCCTTCCGGTGTCACT GAGGTGGTCGTGAAGCTCTTTGACTCTGATCCCATCACTGTGACGGTCCCTGTAGAAGT CTCCAGGAAGAACCCWAAATTTATGGAGACCGTGGCGGAGAAAGCGCTGCAGGAATACC GCAAAAAGCACCGGGAGGAGBJIGACCCAGCTTTCTTGTACAAAGTTGG PXINsect-Dest38clus294T

AACTTTGTACAAAAAAGCAGGCTCGGCCACCATGATGAAGACTCTRCTGYTGTTTGTGG GGCTGCTGCTGACCTGGGAGAGTGGGCAGGTCCTGGGGGACCAAACGGTCTCAGACAAT GAGCTCCAGGAAATGTCCAATCAGGGAAGTAAGTACGTCAATRAGGAAATTCAAAATGC TGTCAACGGGGTGAAACAGATAAAGACTGTGATAGAAAAAACAAACGAAGAGCGCAAGA CACTGCTCAGCAACCTAGAAGAAGCCAAGAAGAAGAAAGAGGATGCCCTAAATGAGACC AGGGAATCAGAGACAAAGCTGAAGGAGCTCCCAGGAGTGTGCAATGAGACCATGATGGC CCTCTGGGAAGAGTGTGAGCSCTGCCTGAGACAGACCTACATGARATTCTACGCACGCG TCTCCACAACTCCCTCACCCCTCMTTCCCCGCCACCTTCACAAATTCCTGAACCACAAC TCRCMCTTCTACTTCTGGATGAATGGTGACCCCATCGACTCCCTGCTCAAAAACCACCC CCACCAGACGCACATACTACATGTCATACAAAACCACTTCAACCGCACATCCAGCATCA TAMCCYACCTCTTCCACGACACGTTCTTCACCCGGAAGCCCCACGCTCCCTACCACTAC CTCCTCTTCAGCCTGCCCCACCAGAAGCCTCACTTCTTCTTTCCCAAGTCCCGCATCGT CCGCAGCTTGATGCCCTTCTCTCCGTACGAGCCCCTGCACTCCCACGCCATGTTCCAGC CCTTCCTTGAGATGATACACGAGGCTCAGCAGGCCATGSACATCCACTTCCACAGCCCG GGCTTCCAGCACCCGCCAACAGKACTCATACGAGAAGGCGACGATGACCGGACTGTGTG CCGGGAGATCCGCCACAACTCCACGGGCTGCTGCGGATGAAGGACCAGTGTGACAAGTG CCGGGAGHTCTTGTCTGTGGACTGATCCACCAAC( |GACCCAGCTMTCTTGTARAAAG TTGG

pXINsect-Dest38clus356T

AAGTTTGTACAAAAAAGCAGGCTCGGCCACC|HATGAAGACTCTGCTGCTGTTTGTGG GGCTGCTGCTGACCTGGGAGAGTGGGCAGGTCCTGGGGGACCAAACGGTCTCAGACAAT GAGCTCCAGGAAATGTCCAATCAGGGAAGTAAGTACGTCAATAAGGAAATTCAAAATGC TGTCAACGGGGTGAAACAGATAAAGACTCTCATAGAAAAAACAAACGAAGAGCGCAAGA CACTGCTCAGCAACCTAGAAGAAGCCAAGAAGAAGAAAGAGGATGCCCTAAATGAGACC AGGGAATCAGAGACAAAGCTGAAGGAGCTCCCAGGAGTGTGCAATGAGACCATGATGGC CCTCTGGGAAGAGTGTAAGCCCTGCCTGAMACAGACCTGCATGAAGTTCTACGCACGCG GTCTGCAGAAGTGGGTCAGGCCTGGTTGGCCGCCAGCTTGAGGAGTTCCTGAACCAGAG CTCGCCCTTCTACTTCTGGATGAATGGTGACCGCATCGACTCCCTGCTGGAGAACGACC GGCAGCAGACGCACATGCTGGATGTCATGCAGGACCACTTCAGCCGCGCGTCCAGCATC ATAGACGAGCTCTTCCAGGACAGGTTCTTCACCCGGGAGCCCCAGGATACCTACCACTA CCTGCCCTTCAGCCTGCCCCACCGGAGGCCTCACTTCTTCTTTCCCAAGTCCCGCATCG TCCGCAGCTTGATGCCCTTCTCTCCGTACGAGCCCCTGAACTTCCACGCCATGTTCCAG CCCTTCCTTGAGATGATACACGAGGCTCAGCAGGCCATGGACATCCACTTCCACAGCCC GGCCTTCCAGCACCCGCCAACAGAATTCATACGAGAAGGCGACGATGACCGGACTGTGT GCCGGGAGATCCGCCACAACTCCACGGGCTGCCTGCGGATGAAGGACCAGTGTGACAAG TGCCGGGAGATCTTGTCTGTGGACTGTTCCACCAACAACCCCTCCCAGGCTAAGCTGCG GCGGGAGCTCGACGAATCCCTCCAGGTCGCTGAGAGGTTGACCAGGAAATACAACGAGC TGCTAAAGTCCTACCAGTGGAAGATGCTCAACACCTCCTCCTTGCTGGAGCAGCTGAAC GAGCAGTTTAACTGGGTGTCCCGGCTGGCAAACCTACACGCAAGGCHIGACCCAGCTT TCTTGTACAAAGTTGG PXINsect-Dest38clus403T

AAGTTTGTACAAAAAAGCAGGCTCGGCCACCATdATGAAGACTCTGCTGCRGAWRTGTK GGGCTGCTGCTGACCTGGKAGATTGGGCAGGTCCTGGGGGACCAAACGGTCTCATCCAW TGAGCTCCAGGAAATKTCCAATCAGGGAAGTWAGTACGTCAATAAGGAAATTCAAAATG CTGTCAACGGGGTGAAACATATAAAGACTCTCATAGAAAAAACAAACGAAGAGCGCAAG ACACTGCTCAGCAACCTAGAAGAAGCCAAGAAGAAGATAGAKGATGCCCTATWTTATAC CAGGGAATCATAKACAAAGCTGAAGGAGCTCCCAGGAGTGTGCAATGAKACCATGATGG CCCTCTGGKAAGAKTGTWAGCCCHGCCTGATMYATACCTGCATGAWGTTCTACGCACGC TCCMGCAGAAGAGGCACAGGCCAGGVAGGCCGCCAGCMAGAGGAGAACCAGAACCAGAG CACGCCCAACAACAACAGGAAGAAMGGAGACCGCAACGACACCCAGCAGGAGAACGACC CGGCAGCAGACGCACAAGCAGGAAGACAAGCAGGACCACRACAGCCGCGCGGCCAGCAA CAAAGACGAGCMCCACCAGGAACAGGAACAACACCCGGGAGCCCCAGGAAACCAACCAC AACCMGCCCCACAGCCCGCCCCACCGGAGGCCACACAACAACAAAMCCAGTTCCCGCAT CGTCCGCAGCTTGATGCCCTTCTCTCCGTACGAGCCCCTGAACTTCCACGCCATGTTCC AGCCCTTCKTTGAGTTGATACACGAGGCTCAGCAGGCCATGGACATCCACTTCCACAGC CCGGCCTTCCAGCACCCGCCTTKAGAAATAATACGAGAAGGCGACGATGACCGGACTGT GTGCCGGGAGAWCCGCKACAACTCAACGGGCTGCTGCGGATGAAGGACCWGTGTYACAA GTGCCGGGAGATCTTGTCTGTGGAKTGTTCCACKAACAACKCKTACCAGGCTAAGCTGC GGCGGGYGCTCGAKGAATLCCTCCAGGTAGKTGAGAGGTTGACCAGAAAATACAACGAG CTGCWAAAGTCCTACCAGTGGAAGATGKTCAACACCTCCTCCTAGCTGGAGCAGCTYAA CGAGKAGAAAAACTGGGTGTKCCYGCTAGKAAACATAAAGCAAGGKGAAGACCAGAACT ATCTGCGGGTCACAACGGTGGCTTCCCAKACWTATYAATCGGACGTWCCTTCCGGTGTK AKTGAGGTGGTCGTGAAGCTCTTTGACTKTGATCCCATCACTGTGACGGTCCCTGTAGR AAGTKTCAGG| |GACCCAGCTHTCTTGTACAAAGTTGG

pXINsect-Dest33clusA2 7P

AAGTTTGTACAAAAAAGCAGGCTCGGCCACCATGATGAAGACTCTGCTGCTGTTTGTRG GGCTGCTGCTGACCTGGGAGARTGGGCAGGTCCTGGGGGACCAAACGGTCTCAVACAAT NAVCTCCAGGAAATGTCCAATCAGGGAAGTAAGTACGTCAATAAGGAAATTMAAAATCC CGTCAACGGGGTGAAACAGATAAAGACTCTCATAGAAAWNACAAACGAAGAGCGCAAGA MACTGCTCAGCAACCTAGAAGAAGCCAAGAAGAAGAAAGAGGRTGCCCTAAATRARACC AGGGAATCARAGACAAAGCTGAAGGAGCTCCCAGGAGTGTGCAATGAGACCATGATGGC CCTCTGGRAAGARTGTAAGCMCTGCCTGAVMCMGACCTGCATGARGTTCTACGCWCGCG TCTGCAGAAGTGGCTCAGGCCTGGTTGGCCGCCAGCTTGAGGAGTTCCTGAACCAGAGC TCGCCCTTCTACTTCTGGRTGARTGGTGAVCGCATCGACCTCCCTGCTGGACAACGACC GGCAGCAGACGCACATGCTGGATGTMATGCAGGACCACTTCARCCGCGCGTCCAGCATC AHAVAGMGMRVCTCTTCCAGGACAGGTTCTYCMCCCGGGVAGCCCCAGGATACCTACCA CTACCTGCCCTTCAGCCTGCCCCACCGGAGGCCTCACTTCTTGTTTCCCAAGTCCCGCA TCGTCCGCAGCTTGATGCCTTCTCTCCGTACGAGCCCCTGAACTTCCACGCATGTTCCA GCCCTTCCTTRAACTGATACACGAGGCTCAGCAGGCCATGGACATCCACTTCCACAGCC CGGCCTTCCAAGCACCCGCCAACAGAATTCATACGAGAAGGCGACGATGACCGGACTGT GTGCCGGGAGGCTCAGGCCACTCACGGCTGCTGCCTGTTCCACCAACAACCCCTRCCAG GCTAAGCTGCGGCGGGAGCTCGACGAATCCCTCCAGGTCGCTGAGAGGTTGACCAGGAA ATACAACGAGCTGCTAAAGTCCTACCAGTGGAAGATGCTCAACACCTCCTCCTTGCTGG AGCAGCTGAACGAGCAGTTTAACTGGGTGTCCCGGCTGGCAAACCTCACGCAAGGCGAA GACCAGTACTATCTGCGGGTCACCACGGTGGCTTCCCACACTTCTGACTCGGACGTTCC TTCCGGTGTCACTGAGGTGGTCGTGAAGCTCTTTGACTCTGATCCCATCACTGTGACGG TCCCTGTAGAAGTCTCCAGGAAGAACCCTAAATTTATGGAGACCGTGGCGGAGAAAGCG CTGCAGGAATACCGCAAAAAGCACCGGGAGGAG| (GACCCAGCTTTCTTGTACAAAGT TGG

pXINsect-Dest38clusQ155P

AACTTTGTACAAAAAAGCAGGCTCGGCCACCATGATGAAGACTCTGCTGCTGTTTGTGG GGCTGCTGCTGACCTGGGAGAGTGGGCAGGTCCTGGGGGACCAAACGGTCTCAVACAAT NAVCTCCAGGAAATGTCCAATCAGGGAAGTAAGTACGTCAATAAGGAAATTMAAAATGC GTCAACGGGGTGAAACAGATAAAGACTCTCATAGAAAWNACAAACGAAGAGCGCAAGAIYI ACTGCTCAGCAACCTAGAAGAAGCCAAGAAGAAGAAAGAGGRTGCCCTAAATRARACCA GGGAATCARAGACAAAGCTGAAGGAGCTCCCAGGAGTGTGCAATGAGACCATGATGGCC CTCTGGGAAGAGTGTAAGCSCTGCCTGAGACAGACCTGCATGARGTTCTACGCACGCGT CTGCAGAAGTGGCTCAGGCCTGGTTGGCCGCCAGCTTGAGGAGTTCCTGAACCAGAGCT CGCCCTTCTACTTCTGGRTGARTGGTGACCGCATCGACTCCCTGCTGGAGAACGACCGG CAGCAGACGCACATGCTTGRKGTGRTGCCCGACCACTTCAGCCGCGCGTCCAGCATCAT AGACGAACTCTTCCAGGACAGGTTCTTCACCCGGGAGCCCCAGGATACCTACCACTACC TGCYCTTTCAGCCTGCCCCACCGGAGGCCTCATTTCTTCTTTCCCAAGTCCCGCATCGT CCGCCGCCACAACTCCACGGGCTGCTGCGGATGAAGGACCAGTGTGACAAGTGCCGGGA GATCTTGTCTGTGGACTGTTCCACCAACAACCCCTCCCAGGCTAAGCTGCGGCGGGAGC TCGACCAGCACCCGCCAACAGAATTCATACGYGAAGGCGASGAMGACCGGACTGTGTGC CGGGAGATCCGCTAVAACTCMACGGGCTGCTGCGGWTMAAGGACCAGTGTMARAAGCGG GAGCTCGACGAATCCCTCCAGGTCGCTGAGAGGTTGACCAGMAAATACAACGAGCTGCT WAAGTCCTACCAGTGGAAGATGRTCAACACCTCCTCCTTGCTGGAGCAGCTGWACGTAG CAGTTTAACTGGGTGTCCCGGCTMGCAAACRTCACGCAAGGCGAAGACCAGTACTATCT GCGGGTCACCACGGTGGCTTCCCACACTTCTGACTCGGACGTTCCTTCCGGTGTCACTG AGGTGGTCGTGAAGCTCTTTGACTCTGATCCCATCACTGTGACGGTCCCTGTAGAAGTC TCCAGGAAGAACCCTAAATTTATGGAGACCGTGGCGGAGAAAGCGCTGCAGGAATACCG CWAAAAGCACCGGGAGGAGJ (GACCCTAGCTTTCTTGTACTAATAGTTGG

pXINsect-Dest38clusM228P

AACTTTATACAAAAAAGCAGGCTCGGCCACCI^ATGAAGACTCTGCTGCTGTATGTAG GGCTGCTGCTGACCTGGRAAAATRGGCAGGTCCTGGGGGACCAAMCRGTCTCAAACAAT AARCTCCAGGAAATRTCCAATCAAGGAAGTAAGTACGTCAATAAGGAAATTMAAAATDC TGTCAMCGGGGTGAAACAAATAAARACTCTCATAGAARRRRCRRRCGRRGAGCGCVGGG VGCTGCTCAGCRGCCTRGRGGRGGCCRGGGGGGGGGGGGGGGGTGCCCTGGGTGGGRCC VGGGRGTCTGWGMCYDYGCTGADGGNGCTCCCYGGAGTGTGCGRTGGGGCCRTGGTGGC CCTCTGGGGGGGGTGTGGGCSCTGCCTGGGGSGGGCCTGCGTGRGGTWCTRCGCRCGCG TCTRCAGARGTGGCTCGGGCCTGGTTGGCCGCCAGCDTGAGGGGTTCCTGRGCCAGGGC WCGCCCTTCWRCWTCWGGGTGRGWGGWGGCCGCGTCGRCTCCCTGCTGGAGRGCGAGGC CGTCGTCTCRCACATRCWGGRWRTVGTGCAGGGCCMCTTCRRCCGCGCGTCCCGCAWCA WRRRCGRGCTCTTCCSGGRCAGGTWCTCWCACCCCGGGGGCCCGGGGTACCTACCRCTR CCTACCCTTCGGCCTGCCCCACCGGRGGCCTCACTTCTWCTTTCCCAAGTCCCGCATCG TCCGCAGCTTGATGCCCTTCTCTCCGTACGAGCCCCTGAACTTCCACGCCATGTTCCAG CCCTTCKTTGAGCCCATACACGAGGCTCAGCAGGCCATGGACATCCACTTCCACAGCCC GGCCTTCCAGCACCCGCCAACAGAATTCATACGAGAAGGCGACGATGACCGGACTGTGT GCCGGGAGAWCCGCCACAACTCCACGGGCTGCCTGCGGATGAAGGACCWGTGTGACAAG TGCCGGGAGATCTTGTCTGTGGAKTGTTCCACCAACAACCCCTCCCAGGCTAAGCTGCG GCGGGYGCTCGACGAATCCCTCCAGGTCGKTGAGAGGTTGACCAGAAAATACAACGAGC TGCWAAAGTCCTACCAGTGGAAGATGCTCAACACCTCCTCCTAGCTGGAGCAGCTGAAC GAGKAGTTTAACTGGGTGTKCCGGCTGGKAAACCTCACGCAAGGKGAAGACCAGAACTA TCTGCGGGTCACAACGGTGGCTTCCCAKACWTATYAATCGGACGTWCCTTCCGGTGTKA KTGAGGTGGTCGTGAAGCTCTTTGACTKTGATCCCATCACTGTGACGGTCCCTGTAGRA AGTCCAGGAAGAACCCTAAATTTATGGAGACCGTGGCGGAGAAAGCGCTGCAGGAATAC CGCAAAAAGCACCGGGAGGAG§(GACCCAGCTTTCTTGTARAAAGTTGG

pX!Nsect-Dest38clusL322P

AACTTTGTACAAAAAAGCAGGCTCGGCCACCATGJATGAAGACTCTGCTGCWGAAAWGTR GGGCTGCTGCTGACCTGGGAGAGTGGGCAGGTCCTGGGGGACCAAACGGTCTCAGACAA TRAGCT CCAGGAAATGT C C AAT C AGGGAAGT AAGTACGT C AAT AAGGAAATT CAAAATG CTGTCAACGGGGTGAAACAGATAAAGACTCTCATAGAWAAAACAWACGAAGAGCGCAAG ACACTGCTCAGCAACCTAGAAGAAGCCAAGAAGAAGAAAGAGGATGCCCTAAATGAGAC CAGGGAATCAGAGACAAAGCTGAAGGAGCTCCCAGGAGTGTGCAATGAGACCATGATGG CCCTCTGGGAAGAGTGTAAGCCCTGCCTGAHACARACCTGCATGAAGTTCTACGCACGC GTCTGCAGAAGTGGCTCAGGCCTGGTTGGCCGCCAGCTTGAGGAGTTCCTGAACCAGAG CTCGCCCTTCTACTTCTGGATGADTGGTGACCGCATCGACTCCCTGCTGGAGAACGACC GGCAGCAGACGCACATGCTGGATGTCATGCAGGACCACTTCAGCCGCGCGTCCAGCATC ATAGACGAGCTCTTCCAGGACAGGTTCTTCACCCGGGGAGCCCCAGGATACCTACCACT ACCTGCMCTTCAGCCTGCCCCACGGAGCCTCACTTCTWCTTTCCCCAAGTCCCGCATCG TCCGCAGCTTGATGCCCTTCTCTCCGTACGAGCCCCTGHACTTCCACGCCATGTTCCAS CCCTTCCTTGAGWTGATARACGAGGRTCAGCAGGCRATGMACATCRACTTCCACAGCCC GGCCTTCCAGCACCCGCCWARWGWATTRATACGAGAAGGCGACGATGACCGGACTGTGT GCCGGGAGATCCGCCACAACTCCACGGGCTGCTGCGGATGAAGGACCAGTGTGACAAGT GCCGGGAGWTCTTGTCTGTGGACTGTTCCACCAACAACCCCTCCCAGGCTAAGCTGCGG CGGGAGCTCGACGAATCCCTCCAGGTCGCTGAGAGGTTGACCAGYAAATACAACGAGCC CCTAAAGTCCTACCAGTGGAAGATGCTCAACACCTCCTCCTTGCTGGAGCAGCTGAACG AGCAGTTTAACTGGGTGTCCCGGCTGGCAAACCTCACGCAAGGCGAAGACCAGTACTAT CTGCGGGTCACCACGGTGGCTTCCCACACTTCTGACTCGGACGTTCCTTCCGGTGTCAC TGAGGTGGTCGTGAAGCTCTTTGACTCTGATCCCATCACTGTGACGGTCCCTGTAGAAG TCTCCAGGAAGAACCCTAAATTTATGGAGACCGTGGCGGAGAAAGCGCTGCAGGAATAC CGCAAAAAGCATCCGGGAGGAG§ (GACCCAGCTWTCTTGTACAAAGTTGG

pXINsect-Dest38clusA413P AAGTTTGTACAAAAAAGCAGGCTCGGCCACCf*||fATGAAGACTCTGCTGCTGTTTGTGG GGCTGCTGCTGACCTGGGAGAGTGGGCAGGTCCTGGGGGGACCAAACGGTCTCAGACAA TGAGCTCCAGGAAATGTCCAATCAGGGAAGTAAGTACGTCAATAAGGAAATTCAAAATG CTGTCAACGGGGTGAAACAGATAAAGACTCTCATAGAAAANACAAACGAAGAGCGCAAG ACACTGCTCAGCAACCTAGAAGAAGCCAAGAAGAMGAAAGAGGATGCCCTAAATGAGAC CAGGGAATCAGAGACAAAGCTGAAGGAGCTCCCAGGAGTGTGCAATGAGACCATGATGG CCCTCTGGGAAGAGTGTAAGCCCTGCCTGAAACAGACCTGCATGAAGTMCTACGCACGC GTCTGCAGAAGTGGMTCAGGCCTGGTTGGCCGCCAGCTTGAGGAGTTCCTGAACCAGAG CTCGCCCTTCTACTTCTGGATGARTGGTGACCGCATCGACTCCCTGCTGGAGAACGACC GGCAGCAGACGCACATGCTGGATGTCATGCAGGACCACTTCAGCCGCGCGTCCAGCAMC ATAGACGAGCTCTTCCAGGACAGGTTCTTCACCCGGGAGCCCCAGGATACCTACCACTA CCTGCCCTTCAGCCTGCCCCACCGAGGCCTCACMTCTTCTTTCCCAAGTCCCGCATCGT CCGCAGCTTGATGCCCTTCTCTCCGTACGAGCCCCTGHACTTCCACGCCATGTTCCASC CCTTCCTTGAGWTGATARACGAGGRTCAGCAGGCRATGMACATCRACTTCCACAGCCCG GCCTTCCAGCACCCGCCWARWGWATTRATACGAGAAGGCGACGATGACCGGACTGTGTG CCGGGAGATCCGCCACAACTCCACGGGCTGCTGCGGATGAAGGACCAGTGTGACAAGTG CCGGGAGHTCTTGTCTGTGGACTGTTCCACCHACAACCCCTCCCAGGCTAAGCTGCGGC GGGAGCTCGACGAATCCCTCCAGGTCGCTGAGAGGTTGACCAGGAAATACAACGAGCTG CTAAAGTCCTACCAGTGGAAGATGCTCAACACCTCCTCCTTGCTGGAGCAGCTGAACGA GCAGTTTAACTGGGTGTCCCGGCTGGCAAACCTCACGCAAGGCGAAGACCAGTACTATC TGCGGGTCACCACGGTGGCTTCCCACACTTCTGACTCGGACGTTCCTTCCGGTGTCACT GAGGTGGTCGTGAAGCTCTTTGACTCTGATCCCATCACTGTGACGGTCCCTGTAGAAGT CTCCAGGAAGAACCCTAAATTTATGGAGACCGTGCCCGAGAAAGCGCTGCAGGAATACC GCAAAAAGCACCGGGAGGAGTGAGACCCAGCTVTCTTGTACAAAGTTGG Articles below removed for copyright reasons.

Please refer to published versions of the listed papers:

Poon, S, Easterbrook-Smith, S.B,Rybchyn, M.S., Carver, J.A, and Wilson, M.R. (2000) "Clusterin is an ATP-independent chaperone with very broad substrate specificity that stabilizes stressed proteins in a folding-competent state" Biochemistry: 39, 15953-15960.

Poon, S, Rybchyn, M.S., Easterbrook-Smith, S.B, Carver, J.A, Pankhurst, G.J, and Wilson, M.R. (2002) "Mildly Acidic p H Activates the Extracellular Molecular Chaperone Clusterin" J. Biol. Chem: 277, 39532-39540.

Poon, S , Treweek, T.M, Wilson, M . R , Easterbrook-Smith, S.B, and Carver, J.A.(2002) "Clusterin is an extracellular chaperone that specifically interacts with slowly aggregating proteins on their off-folding pathway" FEBS Letters: 513, 259-266.

Lakins, J.N, Poon, S, Easterbrook-Smith, S.B, Carver, J.A, Tenniswood, M.P.R, and Wilson, M.R. (2002) "Evidence that clusterin has discrete chaperone and ligand binding sites" Biochemistry: 41, 282-291.